ES2246502T3 - TRANSGENIC NON-HUMAN ANIMALS ABLE TO PRODUCE HETEROLOGICAL ANTIBODIES. - Google Patents

Abstract

A transgenic mouse having a genome comprising a human immunoglobulin heavy chain minilocus transgene, the transgene being capable of rearrangement and switching in the mouse, whereby two or more rearranged human heavy chain isotypes are produced.

Description

Transgenic non-human animals capable of produce heterologous antibodies.

Technical field

The invention relates to transgenic mice capable of producing heterologous antibodies. They are also described in this document transgenes used to produce such animals transgenic, immortalized B cells capable of producing heterologous antibodies, methods and vectors to break loci of endogenous immunoglobulins, methods to generate a repertoire of Variable region gene segments of synthetic immunoglobulins and methods to induce the production of heterologous antibodies.

Background of the invention

One of the main impediments facing the development of in vivo applications for monoclonal antibodies in humans is the intrinsic immunogenicity of non-human immunoglobulins. Patients respond to therapeutic doses of rodent monoclonal antibodies by making antibodies against rodent immunoglobulin sequences. These human anti-mouse antibodies (HAMA) neutralize therapeutic antibodies and can cause acute toxicity. The response of HAMA is less dramatic in immunodeficient patients. Therefore, intrinsic immunogenicity has not prevented the use of rodent monoclonal antibodies for the treatment of graft rejection, which implies the temporary attenuation of the patient's immune response. Rodent antibodies may also be useful for treating certain lymphomas that involve immunodeficiencies. However, even immunodeficient patients can create a HAMA response that leads to a reduction in safety and efficacy.

The present technology to generate antibodies monoclonal involves the pre-exposure, or priming, of an animal (usually a rat or mouse) with an antigen. This pre-exposure leads to the formation of B cells splenic secreting immunoglobulin molecules with high affinity for the antigen. Then spleen cells from a spleen are fused animal primed with myeloma cells to form cells of immortal hybridoma, antibody secretors. Are selected individual hybridoma clones to identify the cells that produce immunoglobulins directed against an antigen particular.

Genetic engineering modification of individual antibody genes has been proposed. Two genetic engineering strategies have been reported: chimeric antibodies and transplantation of complementarity determining regions (CDR). The simplest strategy, chimeric antibodies, take advantage of the fact that the variable and constant portions of an antibody molecule are encoded by different exons. By means of a simple fusion of exons of the variable regions of a mouse antibody gene rearranged with exons of a human constant region, a hybrid antibody gene can be obtained (Morrison, SL, et al . (1984), Proc Natl. Acad. Sci. USA , 81 , 6851-6855). The main problem that arises with this strategy is that although the highly immunogenic mouse Fc region is eliminated, the remaining mouse Fab sequences remain immunogenic (Bruggemann, et al . (1989), J. Exp. Med ., 170 , 2153 -2157). The CDR transplant strategy uses a computer model to generate a completely artificial antibody in which the only mouse regions are those involved in antigen binding (Riechmann, L., et al . (1988), Nature , 322 , 323-327). Each of these strategies requires prior characterization of a rodent monoclonal antibody directed against the antigen of interest, and both require the generation of a stable transfected cell line that produces high levels of the genetically engineered antibody.

Another strategy for the production of human antibodies is a proposal that involves the construction of bacterial expression libraries containing immunoglobulin cDNA sequences (Orlandi, et al . (1989), Proc. Natl. Acad. Sci. USA , 86 , 3833 -3837, and Huse, et al . (1989), Science , 246 , 1275-1281). This technique has supposedly only been used to generate antibody fragments derived from mouse cDNA sequences.

Several experiments have reported the use of transfected cell lines to determine the specific DNA sequences required for the reorganization of Ig genes (reviewed by Lewis and Gellert (1989), Cell , 59 , 585-588). These reports have identified probable sequences and have concluded that the accessibility of these sequences to recombinant enzymes used for reorganization is modulated by transcription (Yancopoulos and Alt (1985), Cell , 40 , 271-281). The sequences for the V (D) J junction, it is said, consist of a highly conserved almost palindromic heptamer and a worse preserved AT-rich nanomer separated by a 12 or 23 base pair (bp) spacer (Tonegawa (1983) , Nature , 302 , 575-581; Hesse, et al . (1989), Genes in Dev ., 3 , 1053-1061). Apparently, only effective recombination takes place between sites that contain the recombination signal sequences with spacer regions of different length.

The production of transgenic mice that contain various forms of immunoglobulin genes has also been reported. Heavy chain or light chain genes of reorganized mouse immunoglobulins have been used to produce transgenic mice. These transgenes, it seems, are capable of excluding the reorganization of endogenous Ig genes. See, for example, Weaver et al . (1985), Cell , 42 , 117-127; Iglesias, et al . (1987), Nature , 330 , 482-484; Storb et al . (1985), Banbury Reports , 20 , 197-207; Neuberger et al . (1989), Nature , 338 , 350-352; Hagman et al . (1989), J. Exp. Med ., 169 , 1911-1929; and Storb (1989) in Immunoglobulin Genes, Academic Press, T. Honjo, FW Alt and TH Rabbitts eds. pages 303-326. In addition, functionally rearranged human Ig genes that include the µ or γ-gammal constant region have been expressed in transgenic mice. Yamamura et al . (1986), Proc. Natl Acad. Sci. USA , 83 , 2152-2156; Nussenzweig, et al . (1987), Science , 236 , 816-819. In the case of the reorganized heavy chain gene µ, allelic exclusion of endogenous immunoglobulin gene loci has been reported.

However, allelic exclusion does not always occur in all transgenic B cells. See, for example, Rath, et al , (1989), J. Immunol ., 143 , 2074-2080 (construction of the reorganized gene); Manz, et al . (1988), J. Exp. Med ., 168 , 1363-1381 (transgenes that lacked transmembrane exons did not prevent reorganization of endogenous genes); Ritchie, et al . (1984), Nature , 312 , 517-520 and Storb, et al . (1986), Immunol. Rev., 89, 85-102 (transgenic mice expressing the rearranged κ transgen capable of forming a stable heavy / light chain complex only rearrange endogenous κ genes in B cells that cannot correctly rearrange the heavy chain gene endogenous); and Manz, et al . (1988), J. Exp. Med ., 168 , 1363-1381 (transgenic mice containing the κ gene encoding a light chain unable to combine with heavy chains, show only a low level of allelic exclusion). See also, Nussenzweig, et al . (1988), Nature , 336 , 446-450); Durdik, et al . (1989), Proc. Natl Acad. Sci. USA , 86 , 2346-2350; and Shimizu, et al . (1989), Proc. Natl Acad. Sci. USA , 86 , 8020-8023.

A somatic mutation has also been reported in a 15 kb mouse κ gene construct in hyperimmunized transgenic mice (O'Brien, et al . (1987), Nature , 326 , 405-409; Storb (1989) in Immunoglobulin Genes, Academic Press, T. Honjo, FW Alt, and TH Rabbitts, eds. Pages 303-326) and in the variable portion of the heavy chain transgene (Durdik, et al , (1989), Proc. Natl. Acad Sci. USA, 86 , 2346-2350).

The reorganization of Ig genes, although studied in tissue culture cells, has not been extensively examined in transgenic mice. Only a handful of reports have been published describing reorganization assay constructs introduced in mice [Buchini, et al . (1987), Nature , 326 , 409-411 (γ transgender of unorganized chicken); Goodhart, et al . (1987), Proc. Natl Acad. Sci. USA , 84 , 4229-4233) (non-rearranged mouse κ gene); and Bruggemann, et al . (1989), Proc. Natl Acad. Sci. USA , 86 , 6709-6713 (hybrid mouse-human heavy chain)]. However, the results of these experiments have been variable, producing, in some cases, incomplete or minimal reorganizations of the transgene.

Based on the above, it is clear that there is need for heterologous monoclonal antibodies, for example, antibodies of human origin, derived from a species other than the human In this way, it is an object of the present invention provide a source of monoclonal antibodies that can be used from a therapeutic point of view in the particular species for which have been designed.

According to the previous object, it they provide transgenic mice that are capable of producing a human antibody

B cells can also be provided from of these transgenic mice that are capable of expressing antibody heterologists These B cells can be immortalized to provide a specific monoclonal antibody source for an antigen particular.

According to the above, they can be obtained hybridoma cells that are capable of producing these antibodies Heterologous monoclonal.

The transgenic mice of the invention are produce providing heavy chain transgenes of Immunoglobulin not reorganized heterologists.

In addition, to produce the transgenic mice of the invention methods of breaking loci of endogenous immunoglobulin in transgenic mice.

The production of heterologous antibodies in Transgenic mice mentioned above can be induced by methods described in this document.

One or more suitable transgenes can be constructed for the invention using a repertoire of gene segments of variable region of immunoglobulin generated according to methods described in this document.

Summary of the invention

In accordance with the foregoing objects, the invention provides a transgenic mouse having a genome that includes a transgene comprising DNA sequences encoding the human variable (V), diversity (D), binding (J) and constant regions of a human Ig protein, said sequences being operably linked to transcription regulatory sequences and capable of undergoing a gene reorganization in vivo , when integrated into a non-human transgenic animal, to produce a reorganized gene encoding a heavy chain polypeptide human, said construction also comprising a donor region of change [mu] in position 5 'with respect to a constant region [mu] and an acceptor region of human [gamma change] between the constant region [mu] and a human constant region [gamma], said being change sequences operatively linked to effect change in vivo and for the production of a γ heavy polypeptide chain human In the transgenic mouse B cells, functionally rearranged heterologous immunoglobulin heavy chain transgenes are found.

Immunoglobulin transgenes can be introduced non-rearranged heavy chain and / or light heterologists in a mouse host to produce a transgenic non-human animal that contain a heterologous heavy chain immunoglobulin gene and light or an intermediate animal that contains one or the other transgene. When incorporated into the germ line of these animals intermediate, the crossings between one that contains the chain transgene heavy and other containing the light chain transgene produce transgenic non-human animals that contain the two transgenes of heavy and light chain heterologous immunoglobulin.

The transgenes of the invention include a heavy chain transgene comprising DNA encoding at least one variable gene segment, a diversity gene segment, a binding gene segment and a constant region gene segment. The immunoglobulin light chain transgene comprises DNA that encodes at least one variable gene segment, a gene segment of union and a constant region gene segment. Segments genes that encode the light chain and gene segments of heavy chain are heterologous to the transgenic mouse, as they proceed of, or correspond to a DNA encoding chain gene segments heavy and light human immunoglobulin (that is, they come from a species that is not the transgenic mouse). In accordance with this invention, the transgene is constructed so that the segments individual genetics are not reorganized, that is, they are not rearranged to encode an immunoglobulin heavy chain functional. These unorganized transgenes allow the recombination of gene segments (functional reorganization) and somatic mutation of immunoglobulin heavy chains resulting rearranged within the transgenic mouse when exposes the antigen.

In these transgenic constructions, the different regulatory sequences, for example, sequences promoters, enhancers, class change regions, signals from recombination and the like, comprise corresponding sequences derived from heterologous human DNA. Alternatively, you are regulatory sequences can be incorporated into the transgene from of the same species or of a species related to the mouse used In the invention. For example, gene segments of human immunoglobulin in a transgene with an enhancer sequence of rodent immunoglobulin for use in the transgenic mouse.

Transgenic mice according to the invention can be used to obtain high affinity antibodies as follows. The transgenic mouse that contains immunoglobulin light and heavy chain transgenes no reorganized in the germ line - experiencing VDJ binding during the differentiation of D cells - contacts an antigen to induce the production of a heterologous antibody in a secondary repertoire cell B. This induction causes somatic mutation in rearranged heavy chain transgenes contained in the primary repertoire B cells to produce a heterologous antibody that has high affinity and specificity for the antigen.

These antibody-producing B cells can immortalize by transformation with a virus or with an oncogene that contain a DNA construct, or alternatively, they can immortalized by fusion with a myeloma cell line to forms hybridoma secreting antibodies. In any case, they can select clones that have sufficient affinity and specificity by a particular antigen to provide a source of monoclonal antibody that has low immunogenicity in the species from which the immunoglobulin gene segments of the transgenes (ie, humans).

This document describes vectors and methods to break the endogenous immunoglobulin loci in the animal no human to be used in the invention. These vectors and methods use a transgene, preferably a selection vector positive-negative, which is constructed so that it lead to the functional breakdown of a class of gene segments that encode an endogenous immunoglobulin heavy chain for the mouse using in the invention. These endogenous gene segments include gene segments of the region of diversity, of the region of union and of the constant region. The selection vector positive-negative contacts at least an embryonic mouse stem cell, after which they can select cells in which the vector of positive-negative selection in the mouse genome by homologous recombination. After the transplant, the resulting transgenic mouse is substantially unable to give origin to an immune response mediated by immunoglobulin as result of the homologous integration of the vector. These animals do not humans with immunodeficiency can subsequently be used to studies of immune deficiencies or can be used as receptors of light and heavy immunoglobulin chain transgenes heterologous

Methods for generating a repertoire of synthetic variable region gene segments that are they will use in the transgenes suitable for use in the invention. He method comprises generating a DNA population of segments V of immunoglobulin, where each of the V-segment DNA encodes an immunoglobulin V segment and contains at each end an endonuclease cleavage recognition site of restriction. The population of the V segment DNAs of immunoglobulin subsequently concatenated to form a repertoire of synthetic immunoglobulin V segments.

Transgenic mice can additionally contain heterologous immunoglobulin light chain transgenes Functionally rearranged in the germ line of the animal transgenic These animals contain repertoire B cells primary expressing these light chain transgenes reorganized These B cells are able to undergo mutation somatic when they come into contact with an antigen to form a heterologous antibody with high affinity and specificity for the antigen.

In this way, the invention also includes transgenic mice that contain germline cells with heavy and light chain transgenes, where the rearranged transgene it is an immunoglobulin light chain transgene and the transgene is not rearranged is an immunoglobulin heavy chain transgene.

Methods can be used to produce heterologous antibodies in a transgenic mouse containing cells B of primary repertoire with heavy and light chain transgenes of reorganized heterologous immunoglobulins. These mice transgenic can be obtained from any transgenic mouse previously mentioned. In this way, transgenic mice containing non-rearranged heavy and light chain transgenes, and transgenic mice that contain light chain transgenes reorganized and heavy chain transgenes not reorganized in the germ line of the animal, each containing B cells of primary repertoire that have heavy and light chain transgenes of heterologous immunoglobulin. A first may occur. desired heterologous antibody that is capable of binding to a first antigen. It is known that heavy and light chain transgenes of immunoglobulin reorganized in repertoire B cells Primary of these animals produce repertoire antibodies primary that have sufficient affinity for a second antigen known. In this method, the transgenic mouse can enter contact, sequentially or simultaneously, with the first and the second antigen to induce the production of the first antibody heterologous by somatic mutation of reorganized transgenes. Secondary repertoire B cells produced in this way are then manipulate as previously described to immortalize production of the desired monoclonal antibody capable of binding to First antigen

Plasmids can be used to generate the transgenic mouse according to the present invention. These plasmids are useful for cloning large DNA fragments (e.g., genomic fragments of immunoglobulins), and have an origin of replication (ORI), a copy control region (e.g., ROP, or the copy control region of pACYC177, or others known to those skilled in the art), and a cloning site. Plasmids also include a transcription terminator (eg, trp R or others known to those skilled in the art) downstream of endogenous plasmid derived promoters such as the ampicillin resistance gene ( amp R). . The transcription terminator is located upstream of the cloning site, so that transcripts originating in the engine end upstream of the cloning site. Preferably, the cloning site is flanked by rare restriction sites, which are sites consisting of 7, 8 or more nucleotides, rather than the 6 or less nucleotides that form the most common restriction sites; for example, Not I, Sfi I and Pac I. Uncommon restriction sites also include sites that contain nucleotide sequences that occur infrequently in naturally occurring DNA sequences; that is, with a frequency less than about once every 8,000-10,000 nucleotides.

Brief description of the figures

Figure 1 represents the determining regions of complementarity CDR1, CDR2 and CDR3 and structural regions FR1, FR2, FR3 and FR4 in non-rearranged genomic DNA and in RNA messenger expressed from a heavy chain gene of reorganized immunoglobulin.

Figure 2 represents the chain locus λ human.

Figure 3 represents the chain locus Human kappa.

Figure 4 represents the chain locus heavy human.

Figures 5 and 6 represent the strategy for the generation of a repertoire of synthetic V segments.

Figure 7 represents the breakage strategy Functional endogenous immunoglobulin loci.

Figure 8 represents the secondary response T-cell mediated leading to cell maturation B.

Figure 9 represents the somatic mutation and the clonal expansion of B cells in response to two antigens different.

Figure 10 represents a construction transgenic that contains a rearranged IgM gene linked to a 25 kb fragment containing the constant regions γ3 and human γ1 followed by a 700 bp fragment containing the 3 'enhancer sequence of the rat chain.

Figure 11 is a restriction map of the human κ chain locus representing the fragments to be used to form a light chain transgene by homologous recombination in vivo .

Figure 12 represents the construction of pGP1.

Figure 13 represents the construction of the polylinker contained in pGP1.

Figure 14 represents the fragments used to build a human heavy chain transgene of the invention.

Figure 15 represents the construction of pHIG1 and pCON1.

Figure 16 represents the Cγ1 fragments humans that are inserted into pRE3 (3 'rat enhancer) to form pREG2.

Figure 17 represents the construction of pHIG3 ' and PCON.

Figure 18 represents the fragment containing the segments of the human D region used in the construction of the transgenes of the invention.

Figure 19 represents the construction of pHIG2 (plasmid containing segment D).

Figure 20 represents the fragments that cover the human J \ kappa and human C \ kappa gene segments used in the construction of a transgene of the invention.

Figure 21 represents the structure of pE.

Figure 22 represents the construction of pKapH.

Figures 23A to 23D represent the construction of a positive-negative selection vector for the functional breakdown of the immunoglobulin heavy chain locus endogenous mouse.

Figures 24A to 24C represent the construction of a positive-negative selection vector for the functional breakdown of immunoglobulin light chain loci endogenous in mouse.

Figures 25a to e represent the structure of a vector directed towards the kappa light chain.

Figures 26a to f represent the structure of a vector directed to the mouse heavy chain.

Figure 27 represents the vector map pGPE.

Figure 28 represents the structure of the vector pJM2.

Figure 29 represents the structure of the vector pCOR1.

Figure 31 represents the structure of p γ2.

Figure 32 represents the structure of pVGE1.

Figure 33 represents the test results of human Ig expression in a pHC1 transgenic mouse.

Figure 34 represents the structure of pJCK1.

Figure 35 represents the construction of a variable region of synthetic heavy chain.

Table 1 represents the sequence of the vector pGPe.

Table 2 represents the gene sequence V_ {H} 49.8.

Detailed description

The design of a transgenic mouse that responds to stimulation by foreign antigens with a repertoire of heterologous human antibodies, requires that the transgenes of heterologous immunoglobulin contained in the transgenic animal function properly along the path of development of B cells. As a consequence, transgenes are constructed so that produce one or all of the following points: (1) high level of expression and specific cell type, (2) gene reorganization functional, (3) activation and response to allelic exclusion, (4) expression of a sufficient primary repertoire, (5) transduction of signals, (6) class change, (7) somatic hypermutation, and (8) domain of the transgenic antibody locus during response immune.

As will be apparent from the following description, it is not necessary to meet all the above criteria. For example, in transgenic mice in which endogenous immunoglobulin loci functionally break, the transgene need not activate allelic exclusion. Furthermore, when the transgene comprises a functionally rearranged immunoglobulin light chain gene, the second criterion of functional gene reorganization is unnecessary for that light chain transgene that is already reorganized. As background on molecular immunology, see, Fundamental Immunology , 2nd edition (1989), Paul William E., ed. Raven Press, NY

The Structure and Generation of Antibodies

Immunoglobulins, also known as antibodies, are a group of glycoproteins present in the serum and in tissue fluids of all mammals. They occur in large quantities by plasma cells (also called in this document "secondary repertoire B cells") that is develop from precursor B lymphocytes (referred to in this document "primary repertoire B cells"). These primary repertoire B cells carry a membrane attached to them immunoglobulin that is similar to that produced by the B cells of Secondary repertoire completely differentiated. The contact between primary repertoire B cells and an antigen Strange for induction of antibody formation.

The basic structure of all immunoglobulins is based on a unit consisting of two chains identical light polypeptides and two polypeptide chains identical weights attached through disulfide bridges. Each string lightweight comprises two regions known as the variable region of the light chain and the constant region of the light chain. So similar, the immunoglobulin heavy chain comprises two regions called heavy chain variable region and region heavy chain constant. The constant region of the chain heavy or light is encoded by genomic sequences called Gene segments of heavy or light chain constant region. He use of a particular heavy chain gene segment defines the immunoglobulin class. For example, in humans, the segments constant region genens define the IgM class of antibodies while using a gene segment of the region constant γ, γ2, γ3 or γ4 defines the IgG classes of antibodies as well as IgG subclasses from the IgG1 to IgG4.

The variable regions of the heavy chains and light immunoglobulins together contain the domain of antibody antigen binding. Due to the need for diversity in this region of the antibody to allow binding to a wide range of antigens, the DNA that encodes the region primary or initial repertoire variable comprises several different DNA segments derived from segment families specific variable region genetics. In the case of the region variable light chain, these families comprise segments variable genetics (V) and binding gene segments (J). This way, the initial variable region of the light chain is encoded by a gene segment V and a gene segment J, being each selected among the family of gene segments V and J contained in the genomic DNA of the organism. In the case of the region variable light chain, the DNA encoding the variable region of the initial or primary repertoire of the heavy chain comprises a heavy chain V gene segment, a diversity gene segment of the heavy chain (D) and a gene segment J, being selected each among the families of immunoglobulin gene segments Appropriate V, D and J in genomic DNA.

The Primary Repertoire

The process for the generation of DNA that encodes the genes of the heavy and light chains of the immunoglobulins takes place mainly during the development of B cells. Before the union of the various gene segments of immunoglobulins, gene segments V, D, J and constant (C) are, in most cases, in groups of gene segments V, D, J and C in B cell precursors of primary repertoire Generally, all gene segments of a heavy or light chain are located relatively close in a single chromosome. This genomic DNA before recombination of the various immunoglobulin gene segments is called in this genomic DNA document "not reorganized". During the differentiation of B cells, one of each of the cells is recombined members of the appropriate family of gene segments V, D, J (or only V and J in the case of light chain genes) to form rearranged heavy and light chain immunoglobulin genes of functional form This functional reorganization belongs to the segments. variable region to form DNA that encodes a variable region functional. This process of reorganization of gene segments It seems to be sequential. First the union between D and J of heavy chain, followed by V-to-heavy-chain DJ links and light chain V to J junctions. The DNA that encodes this form initial of a functional variable region in a light chain and / or heavy is called "functionally rearranged DNA" or "DNA reorganized. "In the case of the heavy chain, this DNA is called "reorganized heavy chain DNA" and in the case of light chain, this DNA is called "light chain DNA reorganized. "Similar terminology is used to describe functional rearrangements of the transgenes of the invention.

The recombination of gene segments of the variable region to form variable regions of the functional heavy and light chains is mediated by recombination signal sequences (RSS) flanking competent V, D and J segments from a recombinant point of view. The RSS necessary and sufficient to direct the recombination comprise a symmetric heptamer in dyads, an AT-rich nonamer and an intermediate spacer region of 12 or 23 base pairs. These signals are conserved between different loci and species that perform DJ (or VJ) recombination and are functionally interchangeable. See Oettinger, et al . (1990), Science , 248 , 1517-1523 and references cited in said text. The heptamer comprises the CACAGTG sequence or its analog followed by a non-conserved sequence spacer and after a nonamer having the ACAAAAACC sequence or its analog. These sequences are found in the gene segment J or the downstream side of each gene segment V and D. Again, immediately before the D and J segments of the germ line there are two recombination signal sequences, first the non-number and then again the heptamer separated again by an unconserved sequence. The heptameric and non-American sequences that follow a segment V_ {J}, V_ {H} or D are complementary to those that precede the segments J_ {{}}, D or J_} with which they are to be recombined. The spacers between the heptameric and non-American sequences have a length of twelve base pairs or have a length between 22 and 24 base pairs.

In addition to the reorganization of segments V, D and J, an additional diversity is generated in the primary repertoire of heavy and light chains of immunoglobulin by means of a variable recombination between segments V and J in the light chain and between segments D and J of the heavy chain. This recombination variable is generated by a variation in the exact place in which these segments are joined. This variation in the light chain occurs typically in the last codon of the gene segment V and in the first codon of segment J. Similar imprecision occurs at the junction on the heavy chain chromosome between segments D and J_ {H} and can have an extension of up to 10 nucleotides. In addition, they can insert several nucleotides between gene segments D and J_ {H} and before the gene segments V_ {H} and D that are not encoded by genomic DNA. The addition of these nucleotides is known as diversity of the N region.

After the reorganization of VJ and / or VDJ, the transcription of the reorganized variable region and one or more gene segments of the constant region located downstream of the reorganized variable region produces an RNA transcript primary that after the aid of appropriate RNA has as result an mRNA encoding a light or heavy chain of full length immunoglobulin. These heavy and light chains include a leader signal sequence for secretion to take place and / or insertion of immunoglobulin in the transmembrane region of cell B. The DNA encoding this signal sequence is contained in the first exon of segment V used to form the region variable of the heavy or light chain of immunoglobulin. At MRNAs are also present appropriate regulatory sequences for control mRNA translation to produce polypeptides from heavy and light chain of immunoglobulin encoded which, after proper association with each other, will form a molecule of antibody.

The net effect of these reorganizations on the gene segments of variable region and variable recombination that can take place during this union is the production of a repertoire of primary antibodies. Generally, every B cell that it has differentiated until this state produces a single antibody of primary repertoire During this process of differentiation, they have place cellular events that suppress the functional reorganization of gene segments that are not contained in Ig genes Functionally reorganized. The process by which cells B diploids maintain this mono-specificity is called allelic exclusion.

The secondary repertoire

B cell clones expressing immunoglobulins from the set of sequences comprising the primary repertoire to respond to foreign antigens are immediately available. Due to the limited diversity generated by the simple VJ and VDJ binding, the antibodies produced by the so-called primary response have a relatively low affinity. Two different types of B cells constitute this initial response: the precursors of the primary antibody-forming cells and the precursors of the secondary repertoire B cells (Linton, et al . (1989), Cell , 59 , 1049-1059). The first type of B cells matures producing IgM secretory plasma cells in response to certain antigens. The other B cells respond to the initial exposure to the antigen by entering a T-cell-dependent maturation pathway. It is during this maturation of the T-cell-dependent B cells when a second level of diversity is generated by a process called somatic mutation (also sometimes called hypermutation). These primary repertoire B cells use the immunoglobulin molecules present on their surfaces to bind and internalize the foreign antigen. If the foreign antigen is a protein or is physically bound to another protein antigen, that protein antigen is processed and presented on the cell surface by a major histocompatibility complex (MHC) molecule to an auxiliary T cell which, in turn, induces maturation of B cells. Lanzavecchia (1985), Nature , 314 , 537. This general maturation of B cell is known as the secondary response.

During T cell dependent maturation of B cell clones stimulated by antigen, the structure of the antibody molecule on the cell surface changes two ways: the constant region changes to a non-IgM subtype and the Variable region sequence is modified by substitutions multiples of a single amino acid to produce a molecule of higher affinity antibody. It’s this somatic mutation process, followed by the selection of clones with the highest affinity, which generates highly specific and strongly binding immunoglobulins characterized by the Ig-mediated immune response.

As previously indicated, each region variable of a heavy or light chain of Ig contains a domain of antigen binding. It has been determined by sequencing of amino acids and nucleic acids that somatic mutation during the secondary response takes place through region V, including the 3 complementarity determining regions (CDR1, CDR2 and CDR3) also called hypervariable regions 1, 2 and 3. CDR1 and CDR2 are located within the gene segment variable, while the CDR3 is largely the result of the recombination between V and J gene segments or V, D gene segments and J. The portions of the variable region that do not consist of CDR1, 2 or 3 are usually referred to as structural regions designated FR1, FR2, FR3 and FR4. See Figure 1. During hypermutation, the DNA reorganized muta to give rise to new clones with molecules of Ig altered. Clones with higher affinities for the antigen strangely selectively expand by helper T cells, giving place at affinity maturation of the expressed antibody. The clonal selection typically results in the expression of clones containing the new mutation in the CDR1, 2 and / or regions 3. However, mutations outside these can also occur regions that influence the specificity and affinity of antigen binding domain.

Non-Human Transgenic Animals Capable of Produce Heterologous Antibodies

Transgenic mice of the invention by introducing at least one of the transgenes of immunoglobulin described later in this document in a zygote or an embryo in its early stages of developing a mouse. The mice used in the invention are capable of reorganize immunoglobulin gene segments to produce a primary antibody response that, in addition, are capable of produce a secondary response through somatic mutation of these rearranged Ig genes. The mice are particularly useful since your immune system has been studied a lot, including the Genomic organization of heavy and light chain loci of mouse immunoglobulins. See, for example, Immunoglobulin Genes, Academic Press, T. Honjo, F.W. Alt and T. H. Rabbitts, eds. (1989).

As used in this document, the term "antibody" refers to a glycoprotein comprising the minus two identical light polypeptide chains and two chains identical heavy polypeptides linked by bridges disulfide Each of the heavy and light polypeptide chains it contains a variable region (usually the amino terminal portion of the polypeptide chain) that contains a binding domain that interacts with the antigen. Each of the polypeptide chains heavy or light also comprises a constant region of the polypeptide chains (usually the carboxyl terminal portion) of which some sequences mediate the binding of immunoglobulin to host tissues including various system cells immune, some phagocytic cells and the first component (C1q) of the classic complement system.

As used herein, an "antibody heterologous "is defined in relation to the transgenic organism not human producer of said antibody. It is defined as an antibody which has an amino acid sequence or a DNA sequence coding corresponding to that found in an organism that does not It is the transgenic non-human animal. In this way, before the reorganization of a transgene that contains various segments heavy or light chain genetics, these gene segments can be easily identified, for example, by hybridization or DNA sequencing, as coming from a kind of organism different from that of the transgenic animal. For example, according to the present invention, various genome gene segments are used human in heavy chain transgenes in an unorganized form. These transgenes are introduced into mice. Gene segments do not rearranged heavy chain transgene have DNA sequences unique to the human species that are distinguishable from the segments endogenous immunoglobulin gene in the mouse genome. They can easily detected in its unorganized form on the line germinal and in somatic cells that are not B cells and in their form reorganized into B cells.

Certain specific segments of transgenes from reorganized immunoglobulin light chain corresponding to VJ segments reorganized functionally, contain sequences of immunoglobulin DNA that are also easily distinguishable of the gene segments of endogenous mouse immunoglobulins.

These differences in the DNA sequence also are reflected in the amino acid sequence encoded by said human immunoglobulin transgenes compared to encoded by mouse B cells. In this way, they can human immunoglobulin amino acid sequences detected in non-human transgenic animals of the invention with antibodies specific for immunoglobulin epitopes encoded by gene segments of human immunoglobulin.

Transgenic B cells that contain unorganized transgenes of human origin recombine in a way functional appropriate gene segments to form regions variables of light and heavy chains reorganized so functional. It should be understood that most of the DNA of these reorganized transgenes will not correspond exactly with the DNA sequence of the gene segments from which They obtained these reorganized transgenes. This is because mainly to the variations introduced during the variable recombination and due to mutations introduced by hypermutation during the secondary response. Independently of these modifications in the DNA sequence (as well as in that of amino acids), it will be readily apparent that the antibody encoded by these reorganized transgenes has a sequence of DNA and / or of amino acids that is heterologous to the one normally found in the mouse used for the practice of the invention.

The expression "substantial identity", when refers to polypeptides, indicates that the polypeptide or protein in question it shows at least an identity of approximately 30% with a whole protein of natural origin or a portion thereof, normally an identity of at least about 70%, and preferably an identity of at least about 95%. As used in this document, the expressions "isolated", "substantially pure" and "substantially homogeneous" are used interchangeably in this document and describe a polypeptide protein that has been separated from components that the They accompany naturally. Typically, a monomeric protein is substantially pure when at least about 60 to 75% of a sample presents a single polypeptide skeleton. Little chemical variants or modifications typically share the same polypeptide sequence A substantially pure protein will typically constitute more than about a 85-90% of a protein sample, more commonly approximately 95% and preferably have a purity greater than approximately 99% The purity or homogeneity of the protein it can be indicated in several ways well known in the art, such as polyacrylamide gel electrophoresis of a sample of protein, followed by visualization of a single polypeptide band on a polyacrylamide gel after staining. For certain For purposes, high resolution will be necessary and HPLC or a similar means of purification. A polypeptide lacks substantially of components that associate with it naturally when separated from the native contaminants that accompany it In its natural state. In this way, a polypeptide that is synthesized in a cell system different from the cell from which proceeds naturally will substantially lack the components with which it is naturally associated.

Reorganized Transgenes

As used herein, a "transgene of heavy chain of unorganized immunoglobulin "comprises DNA which encodes at least one variable gene segment, one segment diversity gene, a binding gene segment and a segment constant region gene. Each of the gene segments of said heavy chain transgene derive from, or has a sequence which corresponds to the DNA encoding chain gene segments heavy immunoglobulin of a species that is not the animal not human in which said transgene is introduced. Similarly, as used in this document, a "light chain transgene of non-rearranged immunoglobulin "comprises DNA encoding the minus a variable gene segment, a binding gene segment and at minus a constant region gene segment in which each segment gene of said light chain transgene derives from, or has a sequence corresponding to the DNA encoding gene segments of immunoglobulin light chain of a species that is not the animal non-human in which said chain transgene is introduced light

These heavy and light chain transgenes in this aspect of the invention contain the gene segments previously identified in a non-reorganized manner. This way, between segments V, D and J in the heavy chain transgene and between segments V and J of the light chain transgene is found recombination signal sequences (RSS) interposed appropriate. In addition, these transgenes also include signals from help of appropriate RNA to bind a region gene segment constant to the reorganized variable region VJ or VDJ.

To the extent that the heavy chain transgene it contains more than one gene segment of region C, for example C µ and C γ from the human genome, as explained more forward, "regions of change" are incorporated upstream of each of the gene segments of constant region and downstream of the variable region gene segments to allow for recombination between these constant regions to allow a Immunoglobulin class change, for example, from IgM to IgG. These heavy and light immunoglobulin chain transgenes also contain transcription control sequences that include promoter regions located upstream of the gene segments variable region containing OCTA and TATA motifs.

In addition to promoters, other sequences are used regulators with function mainly in B-line cells. Thus, for example, an enhancer sequence of light chain preferably located between the gene segments J and constant region in the light chain transgene to enhance the expression of the transgene, thus facilitating the exclusion allelic In the case of the heavy chain transgene, it is also They use regulatory enhancers.

Although regulatory control sequences previous promoters and enhancers have been described in a way generic, these regulatory sequences can be heterologous to mouse from the human genomic DNA from which the immunoglobulin gene segments of the heterologous transgene. From alternatively, these regulatory gene segments come from the corresponding regulatory sequences of the mouse genome that Contains the heavy and light chain transgene. These sequences Regulators are used to maximize transcription and translation of the transgene to induce allelic extrusion and to provide relatively high levels of transgenic expression.

According to the invention, each of the immunoglobulin gene segments contained in the transgenes of Heavy chain Ig comes from, or has corresponding sequences to genomic DNA, cDNA or portions thereof of a human being. From this way, these gene segments are heterologous to the mouse in the which is going to introduce the transgene. As a consequence, when these gene segments functionally reorganize and hypermute in the transgenic mouse, the heterologous antibody encoded by these heavy transgenes will have an amino acid sequence and a secondary and general tertiary structure that provides a specific utility against a desired antigen when used Therapeutically in humans. In addition, these antibodies show substantially reduced immunogenicity in comparison with antibodies that are "strange" to the "being human".

The transgenic mice that carry these heavy and light chain transgenes are able to organize a Ig-mediated immune response against a specific antigen administered to the animal B cells are produced within the mouse that They are capable of producing heterologous human antibodies. Behind the immortalization, and the selection of a monoclonal antibody appropriate (Mab), for example, a hybridoma, a source of therapeutic human monoclonal antibody. These Mab humans have a significantly reduced immunogenicity when administered therapeutically to humans.

Examples of antigens that can be used to generate heterologous antibodies in the transgenic animals of the invention containing human immunoglobulin transgenes include bacterial antigens, viral antigens and antigens tumor, as well as particular human B and T cell antigens associated with graft rejection or autoimmunity.

Class change

The use of constant regions \ mu or \ delta mainly determined by an alternative aid, allowing IgM and IgD are coexpressed in a single cell. The other isotypes of heavy chain (γ, α and ε) are only expressed natively after a gene reorganization event that eliminates exons C µ and C δ. This reorganization process gene, called class change, is produced by recombination between the so-called exchange segments located immediately chain above each heavy chain gene (except δ). The individual change segments have a length comprised between 2 and 10 kb, and consist mainly of repeated sequences short The exact point of recombination is different for each individual event of class change.

The ability of a transgenic construct to change isotypes during maturation of B cells has not been directly tested in transgenic mice; however, transgenes should perform this function. Durdik et al . (1989) Proc. Natl Acad. Sci. USA , 86 , 2346-2350) microinjected a reorganized mouse heavy chain gene construct and found that in four independent mouse lines, a high proportion of the transgenic B cells expressed the variable region encoded by the transgene associated with IgG instead of with IgM. Thus, it seems that the isotype change has occurred between the transgene and the endogenous δ constant region present on another chromosome.

As used in this document, therefore, the change sequence expression refers to DNA sequences responsible for the recombination of change. A sequence "donor of change," typically a region of change, will be in position 5 '(that is, chain up) of the region of the construction to be removed during change recombination. The region "acceptor of change" will be among the region of the construction that is removed and the constant replacement region (by example, µ, \ varepsilon, etc.). Since there is no place specific where recombination always takes place, the sequence final gene typically will not be predictable from the building.

The change region (S) of the µ, S gene, is located approximately 1 to 2 kb in 5 'position with respect to the coding sequence and is composed of numerous tandem repeats of sequences of the form (GAGCT) n} (GGGGT), where n is normally 2 to 5, but may vary up to 17. ( See T. Nikaido, et al . (1981): Nature , 292 : 845-848).

Similar internally repeated sequences of change have been found that span several 5'-kilobases of the other CH genes. The S? Region has been sequenced and has been found to consist of 80 bp homology units repeated in tandem, while S?, S? 2b and S? 3 contain homology units of 49 bp repeated very similar to each other. (See, P. Szurek, et al . (1985): J. Immunol , 135 : 620-626 and T. Nikaido, et al (1982): J. Biol. Chem ., 257 : 7322-7329). All sequenced S regions include numerous occurrences of the GAGCT and GGGGT pentamers which are the repeated basic elements of the S µ gene (T. Nikaido, et al . (1982): J. Biol. Chem ., 257 : 7322- 7329); in the other regions S these pentamers are not repeated in tandem as accurately as in S, but instead are included in larger repeated units.

The Sγ1 region has an additional structure of greater order: two direct repeated sequences surround each of two groups of 49 bp tandem repeats. ( See MR Mowatt, et al (1986): J. Immunol., 136: 2674-2683). Regions of human H-chain gene change have been found very similar to their mouse counterparts. Generally, unlike what happens with the VJ recombination enzyme machinery, the change machinery can apparently accommodate different alignments of the repeated homologous regions of S germline precursors and then join the sequences at different positions within the alignment. (See, TH Rabbits, et al . (1981): Nucleic Acid Res ., 9 : 4509-4524 and J. Ravetch, et al . (1980): Proc. Natl. Acad. Sci. USA , 77 : 6734-6738 ).

The induction of class change seems to be associated with sterile transcripts that start upstream of the change segments (Lutzker et al ., 1988 Mol. Cell. Biol ., 8 , 1849; Stavnezer et al . 1988. Proc. Natl. Acad. Sci. USA , 85 , 7704; Esser and Radbruch 1989 EMBO J. , 8 , 483; Berton et al . 1989 Proc. Natl. Acad. Sci. USA , 86 , 2829; Rothman et al . 1990 Int. Immunol . 2 , 621). For example, the observed induction of the sterile γ1 transcript by IL-4 and the inhibition by IFN-γ correlate with the observation that IL-4 promotes class change to γ1 in cultured B cells, while IFN-? Inhibits the expression of γ1. Then, ideally, the transgenic constructs that are intended to undergo a class change should include all cis-acting sequences necessary to regulate these sterile transcripts. An alternative method for obtaining class change in transgenic mice (δ [mu] and [epsilon]) involves the inclusion of 400 bp direct repeat sequences flanking the human [mu] gene (Yasui et al . 1989 Eur J. Immunol ., 19 , 1399). Homologous recombination between these two sequences eliminates the µ gene in B cells that only express IgD.

Monoclonal antibodies

Monoclonal antibodies can be obtained by various techniques with which those skilled in the art will be familiar. In summary, spleen cells of an animal immunized with a desired antigen are immortalized, usually by fusion with a myeloma cell (see Kohler and Milstein, Eur. J. Immunol ., 6 : 511-519 (1976)). Alternative methods of immortalization include transformation with Epstein Barr virus, oncogenes or retroviruses, or other methods well known in the art. Colonies that arise from immortalized single cells are selected for the production of antibodies of specificity and affinity desired by the antigen, and the production of the monoclonal antibodies produced by these cells can be enhanced by various techniques, including injection into the peritoneal cavity of a vertebrate host. In Harlow and Lane, Antibodies: A Laboratory Manual , Cold Spring Harbor, New York (1988), for example, various useful techniques are described in these fields, including: immunization of animals to produce immunoglobulins; monoclonal antibody production; immunoglobulin labeling for use as probes; immunoaffinity purification; and immunoassays.

The Transgenic Primary Repertoire A. The Human Immunoglobulin Loci

An important requirement for the function transgenic is the generation of an antibody repertoire primary that is diverse enough to trigger a Secondary immune response against a wide range of antigens. The size of the human immunoglobulin loci encoding the various gene segments of the heavy and light chains is pretty big. For example, in the human genome, the three loci separated for the λ light chain locus, the locus of Light chain \ kappa and heavy chain locus together occupy more 5 Mb of DNA or almost 0.2% of the complete genome. Each locus consists of multiple variable segments that recombine during the B cell development with a segment of binding region (and the heavy chain locus with diversity region segments) to form exons from the entire V region. These light chain genes Reorganized consist of 3 exons: an exon signal peptide, an exon of variable region and a constant region exon. The chain gene Heavy rearranged is somewhat more complex. It consists of an exon of signal peptide, a variable region exon and a tandem series of constant region regions of multiple domains, being each one encoded by several exons. Each of the constant region genes encodes the constant portion of a class of different immunoglobulins. During cell development B, constant regions near region V are removed leading to the expression of new kinds of heavy chain. For each kind of heavy chain, the help of alternative patterns of RNA gives rise to both transmembrane immunoglobulins and secreted immunoglobulins.

Approximately 40% of human serum antibody molecules contain λ light chains. The structure of this locus, which is located on chromosome 22, is the least characterized (fig. 2). It consists of an unknown number of segments V upstream of a tandem series of 6 constant region genes, each of which is linked to a single J segment. In addition, two more constant region segments have been isolated with associated J segments. , although their union with the rest of the λ group has not been established and it is not known whether they are used. E. Selsing, et al ., "Immunoglobulin Genes", Academic Press, T. Honjo, FW Alt and TH Rabbitts, eds. (1989).

The light chain locus \ kappa extends on three groups on chromosome 2 (fig. 3). The first two groups, which cover 850 and 250 kb respectively, contain only gene segments of variable region. The third group, which covers approximately 1 Mb, contains approximately 40 gene segments V string up a group of 5 segments J followed by a single gene segment of constant region. A total of 84 gene segments V, and it is believed that approximately half of they are pseudogenes (Zachau (1989) in Immunoglobulin Genes, Academic Press, T. Honjo, F. W. Alt and T.H. Rabbitts, eds. pages 91-110).

Approximately 25 kb downstream of the region CK there is a "deletion element \ kappa" (\ kappade). The κ recombinant sequence with sequences that are chain above, causing the deletion of the constant region \ kappa in B cells expressing the λ light chain. This results in Isotypic exclusion in cells that rearrange satisfactory both κ and λ genes.

The human heavy chain locus is the longest and most diverse. It consists of approximately 200 V gene segments that span 2 Mb, approximately 30 D gene segments that span approximately 40 kb, 6 J segments grouped within a 3 kb extension and 9 constant region gene segments that extend over approximately 300 kb. The complete locus covers approximately 2.5 Mb of the distal portion of the long arm of chromosome 14 (fig. 4). Heavy chain segments V can be grouped into 6 families based on sequence similarity. There are approximately 60 family members V_ {H} 1, 30 segments V_ {H} 2, 80 segments V_ {H} 3, 30 segments V_ {H} 1, 3 segments V_ {H} 5 and one segment V_ {H } 6. Berman, JE, et al . (1988), EMBO J. , 7 , 727-738. In the human heavy chain locus, the members of the individual V families are intermingled, unlike what happens in the mouse locus where the related V segments are grouped. The only member of the VH family 6 is the closest to the V segments, being located within 90 kb of the constant region gene segments. Sato, T., et al . (1988), Biochem. Biophys Res. Comm ., 154 , 265-271. It seems that all functional segments D and J are in this 90 kb region (Siebenlist, et al . (1981), Nature , 294 , 631-635; Matsuda, et al . (1988), EMBO J. , 7 , 1047-1051; Buluwela, et al . (1988), EMBO J. , 7 , 2003-2010; Ichihara, et al . (1988), EMBO J. , 7 , 4141-4150; Berman, et al . (1988) , EMBO J. , 7 , 727-738).

B. Transgenes of Gene Fragments 1. Heavy Chain Transgen

According to the invention, transgenic mice have a genome that includes or an immunoglobulin heavy chain transgene comprising non-rearranged genomic DNA from humans. Preferably, the heavy chain transgene comprises a NotI fragment having a length between 670 and 830 kb. The length of this fragment is ambiguous because the restriction site 3 'has not been accurately mapped. However, it is known to be located between the? 1 and? Ps? Gene segments (see Figure 4). This fragment contains members of the 6 known VH families, the gene segments D and J, as well as the constant regions µ, δ, γ3, γ1 and α1. Berman, et al . (1988), EMBO J., 7, 727-738. A transgenic mouse line containing this transgene correctly expresses all the heavy chain classes required for the development of B cells as well as a sufficiently large repertoire of variable regions to trigger a secondary response for most antigens.

2. Light Chain Transgen

Similarly, a fragment can be constructed genomic that contains all the necessary gene segments and the regulatory sequences of a human light chain locus. This Construction is described in the examples.

C. Intracellularly Generated Transgenes by In Vivo Recombination

It is not necessary to isolate all or part of the heavy chain locus in a single DNA fragment. Thus, for example, the 670-830 kb NotI fragment of the human immunoglobulin heavy chain locus can be formed in vivo in the mouse during transgenesis. This in vivo transgenic construction is produced by introducing two or more overlapping DNA fragments into an embryonic nucleus of the non-human animal. The overlapping portions of the DNA fragments have DNA sequences that are substantially homologous. After exposure to the recombinases contained in the embryonic nucleus, the overlapping DNA fragments are homologously recombined in the proper orientation to form the 670-830 kb NotI heavy chain fragment.

However, it should be understood that the in vivo transgenic construct can be used to form any number of immunoglobulin transgenes that due to their size are difficult or impossible to obtain otherwise or manipulate by the present technology. Thus, the construction of in vivo transgenes is useful for generating immunoglobulin transgenes that are larger than DNA fragments that can be manipulated by YAC vectors (Murray and Szostak (1983), Nature , 305 , 189-193). This in vivo transgene construct can be used to introduce into a non-human animal substantially the complete immunoglobulin loci from a species that is not the non-human transgenic animal. Thus, although several groups have successfully constructed libraries containing 50-200 kb DNA fragments in YAC vectors (Burke, et al . (1987), Science , 236 , 806-812; Traver, et al . (1989) , Proc. Natl. Acad. Sci. USA , 86 , 5898-5902) and have used polyamine condensation to produce YAC libraries with a size ranging from 200 to about 1000 kb (McCormick, et al . (1989), Proc. Natl. Acad. Sci. USA , 86 , 9991-9995), it is expected that multiple overlapping fragments that substantially cover more than the 670-830 kb NotI fragment of human immunoglobulin constant region loci readily produce major transgenes through methods described in this document.

In addition to forming genomic immunoglobulin transgenes, homologous recombination in vivo can also be used to form "minilocus" transgenes as described in the examples.

When the in vivo transgenic construct is used, each overlapping DNA fragment preferably has a substantially overlapping homologous DNA sequence between the final portion of a DNA fragment and the final portion of a second DNA fragment. These overlapping portions of the DNA fragments preferably comprise from about 500 bp to about 2000 bp, more preferably from 1.0 kb to 2.0 kb. Homologous recombination of overlapping DNA fragments to form transgenes in vivo is also described in the commonly assigned US Patent Application entitled "Intracellular Generation of DNA by Homologous Recombination of DNA Fragments"("Intracellular Generation of Homologous Recombination of Fragments" of DNA ") filed on August 29, 1990, under USSN 07 / 574,747.

D. Minilocus transgenes

As used herein, the expression "immunoglobulin minilocus" refers to a sequence of DNA (which may be within a larger sequence), usually of less than about 150 kb, typically with a length between approximately 25 and 100 kb, which contains at least one of each of the following: a variable gene segment (V) functional, a functional region of union region (J), a segment constant region (C) functional gene, and if it is a minilocus of the heavy chain, a region of functional diversity (D) region, of so that said DNA sequence contains at least one discontinuity substantial (for example, a deletion, usually of at least approximately 2 to 5 kb, preferably 10-25 kb or more, with respect to the homologous genomic DNA sequence). A Light chain minilocus transgen will be at least 25 kb of length, typically 50 to 60 kb. A heavy chain transgene typically it will be approximately 70 to 80 kb in length, preferably at least about 60 kb, with two regions constants operatively linked to regions of change, compared to at least about 30 kb with a single constant region and regions of incomplete change. In addition, the individual elements of the minilocus are preferably in the line configuration germinal and may undergo gene reorganization in cells pre-B of a transgenic animal to express functional antibody molecules with various specificities of antigen completely encoded by the elements of the minilocus

In accordance with the present invention, the immunoglobulin heavy chain transgenes used comprise one or more of each of the following gene segments: V, D, J and C. In the minilocus transgen at least one of each is incorporated gene segment of the appropriate type. With respect to segments C for the heavy chain transgene, the transgene contains at least one gene segment [mu] and at least one gene segment [lambda], plus preferably λ3 or λ1. This preference is for allow class switching between the IgM and IgG forms of the immunoglobulin encoded to provide somatic mutation and the production of a secretable form of high non-IgM immunoglobulin affinity. Other region gene segments can also be used. constant such as those coding for the production of IgD, IgA and IgE.

The heavy chain J region segments in humans comprise 6 functional J segments and 3 pseudogenes grouped in a 3 kb DNA stretch. Given their relatively compact size and the ability to isolate these segments together with the µ gene and the 5 'portion of the δ gene in a single 23 kb SFiI / SpeI fragment (Sado, et al . (1988), Biochem. Biophys Res. Comm ., 154 , 264271), it is preferable to use all gene segments of the J region in the construction of the mini-locus. Since this fragment encompasses the region between the γ and δ genes, it is likely to contain all 3 'cis-linked regulatory elements required for the expression of µ. In addition, since this fragment includes the entire J region, it contains the heavy chain enhancer and the change region (Mills, et al . (1983), Nature , 306 , 809; Yancopoulos and Alt (1986), Ann. Rev Immunol ., 4 , 339-368). It also contains the transcription start sites that trigger VDJ binding to form primary repertoire B cells (Yancopoulos and Alt (1985), Cell , 40 , 271-281). Alternatively, a 36 kb BssHII / SpeIl fragment, which includes part of the D region, may be used instead of the 23 kb SfiI / SpeIl fragment. The use of this fragment increases the amount of 5 'flanking sequences to facilitate the efficient binding of D to J.

The human D region consists of 4 to 5 9 kb homologous subregions, joined in tandem (Siebenlist, et al . (1981), Nature , 294 , 631-635). Each subregion contains up to 10 individual D segments. Some of these segments have been located on the chromosomal map and are shown in fig. 4. Two different strategies are used to generate a minilocus D region. The first strategy involves using only the D segments located in a contiguous short stretch of DNA that includes one or two of the repeated D subregions. A candidate is a single 15 kb fragment that contains 12 individual D segments. This piece of DNA consists of 2 contiguous EcoRI fragments and has been completely sequenced (Ichihara, et al . (1988), EMBO J. , 7 , 4141-4150). Twelve segments D should be sufficient for a primary repertoire. However, given the dispersed nature of the D region, an alternative strategy is to jointly link several fragments containing the non-contiguous segment D to produce a smaller piece of DNA with a larger number of segments.

To construct the heavy chain minilocus transgene at least one is used, and preferably more than one V gene segment. A 10-15 kb DNA piece containing one or two V segments not reorganized together with flanking sequences is isolated. A clone containing this DNA is selected using a probe generated from unique 5 'sequences determined from the transcribed V region of a human hybridoma characterized so as to produce an anti-cytomegalovirus antibody (Newkirk et al . (1988) J Clin. Invest ., 81 , 1511-1518). The 5 'untranslated sequence of the heavy chain mRNA is used to construct a single nucleotide probe (preferably approximately 40 nucleotides in length) to isolate segment V of the original germ line that generated this antibody. The use of a segment V that is known to be incorporated into an antibody against a known antigen not only ensures that this segment V is functional, but also helps in the analysis of the participation of the transgene in secondary immune responses. This segment V is fused with the minilocus region D and constant region fragments, previously described, to produce a heavy chain minilocus transgene.

Alternatively, a contiguous and large stretch of DNA containing multiple segments of region V can be isolated from a YAC library. In DNA pieces of different sizes containing different numbers of V region segments, the ability to provide a repertoire of human antibodies in the construction of the minilocus transgene can be tested. It is also possible to construct a large fragment from several fragments containing non-contiguous V segments using YAC vectors (Murray and Szostak (1983), Nature , 305 , 189-193), plasmids based on factor F (O'Conner, et al . (1989), Science , 244 , 1307-1312) or the in vivo construction mentioned above using recombination of overlapping fragments. Alternatively, a synthetic region V repertoire can be used (described below in this document).

Similarly, a transgene can be constructed of light chain minilocus from the immunoglobulin locus λ or human κ. The construction of a minilocus of light chain \ kappa is very similar to the construction of the heavy chain minilocus, except because it is much simpler due to its smaller size and its lower complexity. The locus \ kappa human contains only a constant region segment, and this segment, together with the 5 'and 3' enhancers and the five segments J functional, can be isolated in a single 10 kb DNA fragment. This fragment is injected together with a V region of minilocus built as described for the chain minilocus heavy

In this way, for example, a construction of a heavy chain minilocus transgen of immunoglobulin, for example, approximately 75 kb, which encodes the sequences of regions V, D, J and the constant region, to from a plurality of DNA fragments, of which at least 2, 3 or 4 are a sequence of the V region, a sequence of the region D, a sequence of regions J and constant, a sequence of regions D, J and constant or a sequence of the region constant, each sequence being substantially homologous to human gene sequences The sequences are linked together operational to transcription regulatory sequences and can undergo reorganization There is also recombination of change with two or more constant region sequences (i.e., µ and γ) appropriately located and regions of change. A exemplary light chain transgenic construction shaped similar from a plurality of DNA fragments, substantially homologous to human DNA and capable of experiencing reorganization, will include at least two, three or four fragments of DNA encoding the V, D and constant regions, each comprising fragment a sequence of region V, a sequence of region J and of the constant region, or a sequence of the region constant.

E. Methods for Determining Gene Segments V Functional and to generate a repertoire of Segments V Synthetic

Of the various families of gene segments, that is, gene segments of regions V, D, J and C, the number of gene segments V generally far exceeds the number of corresponding gene segments for the gene segments of regions D, J and C. By analogy with the rabbit system in which a single V gene segment is used for approximately 90% of the antibodies produced (Knight and Becker (1990), Cell , 60 , 963-970), it is possible to produce transgenes heavy and light containing a limited number of gene segments of the V region, and nothing more than a gene segment of the V region. Therefore, it is desirable to have a method to determine which gene segments of the V region are used in a particular organism, such as the human being, when an immune response mediated by immunoglobulin is generated. According to this strategy, a single gene segment V, when combined with the gene segments J or DJ, is capable of providing sufficient diversity in CDR3 for the generation of a primary repertoire that, after a somatic mutation, can provide additional diversity along the variable region, for example, in CDR1 and CDR2, for the production of high affinity antibodies.

In this way, methods and vectors can be used to determine which V gene segments are commonly used in a organism during an immune response. This method is based on the determination of the V segments found in the cDNA synthesized from polyA + RNA from B cells. These methods and vectors can also be used to facilitate the construction of a repertoire of synthetic V segments.

In figures 5 and 6 the profile of this strategy to identify heavy chain V segments and to generate a repertoire of synthetic V segments. This is applicable similarly to identify light chain V segments with an appropriate modification. The first stage is the construction of a cloning vector The preferred starting material is a DNA fragment (approximately 2 kb) that contains a segment V not rearranged together with the 5 'and 3' flanking sequences. This fragment is cloned into a plasmid, such as pGP1 or pGP2 described below containing a polylinker site flanked by rare cut restriction sites denominated "w" and "z" in figures 5 and 6 (the polylinkers and the restriction sites of pGP1 and pGP2 are described in the examples). Then mutagenesis directed to oligonucleotides to introduce two new restriction sites "x" e "y" (generally, each of approximately 6 nucleotides in length). The restriction site "x" is placed at approximately 20 nucleotides from the 3 'end of the intron, between the signal and exon of segment V. The restriction site "y" is puts approximately 20 nucleotides in the 3 'direction of the junction with segment V, within the 23 bp spacer between the Signal sequences of heptameric and non-American recombination. Cutting up the resulting plasmid with the enzymes "x" e "y" is removed the second exon (segment V), leaving the flanking sequences 5 ', the promoter of the V region, the exon of the signal peptide, the intron, a hollow flanked by "x" and "y" ends, the outer half of the recombination signal sequence and the 3 'flanking sequences. This plasmid is called pVH1.

The second stage is the synthesis of four oligonucleotide primer sets, from P1 to P4. P1 and P2 are non-unique oligomers that have approximately 50 nucleotides each one and that are used to prime the synthesis of double stranded cDNA. P1 start (going from 5 'to 3') with approximately 20 nucleotides of homologous sequence to the antisense chain of the signal sequence of recombination in pVH1 (including the recognition sequence of the restriction enzyme "y"), and continues with approximately 30 antisense sequence nucleotides that hybridize with approximately the last 30 nucleotides of the structural region 3 of VH (FR3). Random bases are incorporated in approximately last 30 nucleotides to thereby generate a set of primers that hybridize with all the different families of VH. He second oligonucleotide, P2, is in the sense orientation, and is homologous to approximately 50 nucleotides starting from the restriction site "x" in pVH1. This includes the site of "x" restriction, approximately the last 20 nucleotides of the intron and approximately the first 30 nucleotides of FR1. From again, approximately the last 30 nucleotides are not unique, to accommodate different segments of the VH region. The oligonucleotides P3 and P4 are homologous to approximately first 20 nucleotides of P1 and P2, respectively. These oligos they are unique, so that they prevent the introduction of new mutations in the V segments and are used to amplify the cDNA of double chain by means of polymerase chain reaction (PCR)

The 3 'terminal portions of primers P1 and P2 that are capable of hybridizing and priming the synthesis of variable segments of immunoglobulin heavy or light chain locus can be readily determined by one skilled in the art. For example, the nucleotide sequences of several human VH genes have been published, see for example Berman, JE, et al . (1988), EMBO J. , 7 , 727-738 and Kabat, EA, et al . (1987), Sequences of Protein of Immunological Interests , United States Department of Health and Human Services, Washington, DC Similarly, when used to identify and / or generate V segments of the human immunoglobulin light chain locus, portions appropriate sequence sequences in the 3 'position of primers P1 and P2 can easily be determined from the published sequences. See, for example, Kabat, EA, et al ., Supra . In general, nucleotide positions that are conserved between various segments V are also conserved in the 3 'portion of primers P1 and P2. For the nucleotide positions in which variation between variable segments is observed, these nucleotide positions in the corresponding primers P1 and P2 have been similarly varied to provide primers P1 and P2 comprising a group of primers capable of hybridizing with different VH segments or VL

The next stage is to use these primers oligonucleotides to generate a cDNA sequence library of the human heavy chain region V in the pHV1 vector. P1 is used to prime the synthesis of the first cDNA chain from RNA human B cell polyA +. The RNA undergoes basic hydrolysis, and the synthesis of the second chain is primed with P2. Then it is purified the full length double stranded cDNA in a gel acrylamide, electroeluted and used as a template for amplification by polymerase chain reaction (PCR) using oligonucleotide primers P3 and P4. So Alternatively, the cDNA is first synthesized by conventional methods and this cDNA is used as a template for the reactor primed with P1. Then, the amplified product (approximately 0.3 kb) is gel purifies, cleaves with restriction enzymes "x" e "y" and is cloned into pVH1.

The resulting cDNA library represents a synthetic genomic library of variable region segments and offers three advantages over a conventional genomic library of Variable segments First, this library does not contain pseudogenes, while a conventional library could contain up to 50% of pseudogenes sequences. In second instead, the synthetic library is more compact than a library conventional, containing a functional V segment for every 2 kb of DNA, instead of a functional segment for every 20 kb. Finally, this strategy leaves the promoter sequences of segment V accessible for handling.

This cDNA library can be skewed towards particular V segments of the germ line due to the expression differential. The two sources of bias are: (i) proportions segment V recombination differentials, and (ii) selection differential of B cell clones expressing segment V. The First source of bias is treated in two ways. First, it removes fetal tissue as a source of B-cell RNA, since the bias is more pronounced in the immunoglobulin repertoire fetal Second, the semi-random primers, P1 and P2, are they are divided into groups, of which each one presents cross hybridization selectively with different families of segment V. Then, these primers are used to generate 4 to 6 libraries separated, thus ensuring that all the families of region V. The second source of bias, the selection B cell clone differential, it is also treated in two ways analogous First, an RNA source that includes the minimum fraction of B cells selected by antigen. Are avoided lymph nodes and spleen. The adult bone marrow is a source of unselected B cells. However, it may contain a high proportion of pseudogenes sequences transcribed to from pre-B cells. Another source of RNA is the whole blood Ninety percent of circulating B cells they are immature cells that express o or mu and δ, and have recently migrated to the bone marrow. However, the level of cells expressing IgG selected by antigen may vary depending on the immune status of the individual. Therefore, in the RNA isolated polyA + B cell sequences are checked selected by hybridization of Northern blot with specific probes In case it is more practical to use RNA from spleen, and if this RNA contains a high fraction of IgG sequences, a second strategy is used to minimize the bias of the selection. The synthesis of the first cDNA chain is primed with a approximately 40 nucleotide primer of exon 2 of the region constant that is specific for IgM transcripts. Then the synthesis of the second chain is primed with P2, and a third round Synthesis is primed with P1. The cDNA of this third round of synthesis provides the template for PCR amplification using P3 and P4.

Once the library has been generated variable region, the V segments used in the present invention by conventional techniques, for example, by sequencing and / or hybridization medium with oligonucleotides with family or segment specificity, as well as by amplification differential by PCR methods. This characterization of the V segment library provides information on the frequency and distribution of the use of V segments in a particular agency and, as a consequence, the identification of V segments that can be used in the construction of the various transgenes of the invention. In this way, in the construction of minilocus transgen described above can be used one or more predominant V gene segments. Additionally, the clones selected from this library can be used to identify genomic fragments containing used V segments frequently to facilitate fragment identification genomics containing a particular desired segment V.

In addition, a repertoire of synthetic V segments by concatenating the sequences of the library. In the production of transgenic mice commonly generate large tandem series of repeated transgenes that They contain hundreds of copies of the injected sequence. These series in tandem they are usually quite stable. However, for ensure the stability of the synthetic V region, preferably introduce random DNA blocks between each segment of region V 2 kb These random DNA blocks are prepared by digesting and then releasing genomic DNA, to prevent the insertion of dominant regulatory elements. Genomic DNA preferably it is digested with four restriction enzymes of frequent cut: AluI, DpnI, HaeIII and RsaI. This digestion produces fragments of blunt ends with an average length of 64 nucleotides Fragments are eluted in the size range of 50 to 100 nucleotides of an acrylamide gel and are religed. The DNA religado is partially digested with MboI and is divided by sizes. Fragments in the range of 0.5 to 2 kb are cloned into the BamHI or BglII site of the vector polylinker used for generate pVH1.

The random sequence library is combined with the library of synthetic V segments to create a repertoire of synthetic V segments. The inserts from the random sequence library are released with the "w" and "z" enzymes, and purified from vector sequences. Inserts from the library of synthetic V segments are isolated by cutting with "w" and "z". Before purifying the V segment inserts, this DNA is treated with calf intestinal phosphatase, to prevent self-ligation. The V segment inserts are then linked together with the random inserts to generate an alternative tandem series comprising a repertoire of synthetic V segments. This ligation mixture is selected by sizes in a sucrose gradient, and the 50-100 kb fraction
It is microinjected together with, for example, a DJ-constant microlocus construction. By directly injecting the repertoire of synthetic V segments without the involvement of a cloning stage, it is possible to take advantage of the fact that the tandem series of injected fragments are inserted into a single site. In this case, these tandem series are not completely redundant, but lead to additional diversity. Alternatively, the repertoire of synthetic V segments can be combined with a DJC minilocus to form a heavy chain transgene.

Similarly, a repertoire can be constructed of synthetic immunoglobulin light chain segments using the appropriate primers for the light chain locus.

Functional Breakdown of Endogenous Immunoglobulin Loci

It is expected that the expression of transgenes from heavy and light chain of reorganized immunoglobulins satisfactorily have a dominant effect through the suppression of the reorganization of immunoglobulin genes endogenous in the non-human transgenic animal. However, another way to generate a non-human animal lacking endogenous antibodies is by mutation of endogenous immunoglobulin loci. Using the embryonic stem cell technology and homologous recombination, you can easily remove the immunoglobulin repertoire endogenous The functional description of mouse immunoglobulin loci. Vectors and methods described, however, can be easily adapted for use in Other non-human animals.

In summary, this technology involves the inactivation of a gene, by homologous recombination, in a pluripotent cell line that is able to differentiate into germ cell tissue. A DNA construct that contains an altered copy of a mouse immunoglobulin gene is introduced into the embryonic stem cell nuclei. In a portion of the cells, the introduced DNA is recombined with the endogenous copy of the mouse gene, replacing it with the altered copy. Cells that contain the newly engineered genetic lesion are injected into a host mouse embryo, which is reimplanted into a recipient female. Some of these embryos develop chimeric mice that possess germ cells completely derived from the mutant cell line. Thus, by crossing the chimeric mice it is possible to obtain a new line of mice containing the introduced genetic lesion (reviewed by Capecchi (1989), Science , 244 , 1288-1292).

Because the mouse \ lambda locus contributes only to 5% of immunoglobulins, the inactivation of heavy chain and / or light chain loci \kappa. There are three ways to break each of these loci: J region deletion, intron enhancer deletion J-C and breakage of coding sequences in the region constant by the introduction of a termination codon (codon stop) The last option is the most direct in terms of the design of DNA constructs The elimination of the gene interrupts the maturation of B cells, thus preventing the change of class to any of the functional heavy chain segments. The Strategy to cancel these loci is summarized below.

To break the mouse µ and κ genes, direction vectors based on the design employed by Jaenisch et al . (Zijlstra, et al ., (1989) Nature , 342 , 435-438) have been used for successful disruption of the mouse β2-microglobulin gene. The neomycin (neo) resistance gene of plasmid pMCIneo is inserted into the coding region of the target gene. The insertion of pMCIneo uses a hybrid promoter / enhancer viral sequence to direct neo expression. This promoter is active in embryonic stem cells. Therefore, neo can be used as a selectable marker for the integration of the knock-out construction. The HSV thymidine kinase (tk) gene is added to the end of the construct as a negative selection marker against random insertion events (Zijlstra, et al ., Supra ).

The direction vectors for the breakage of the Heavy chain locus are illustrated in Figure 7. The strategy Primary for heavy chain locus rupture is elimination from the J region. This region is quite compact in the mouse, covering only 1.3 kb. To build an address vector to a gene, a 15 kb KpnI fragment is isolated that contains all constant region exons A secreted from a library mouse genomics The 1.3 kb J region is replaced by the insert 1.1 kb of pMCIneo. Then the HSV tk gene is added in the 5 'end of the KpnI fragment. The correct integration of this construction, by means of homologous recombination, will lead to the replacement of the mouse JH region with the neo gene (figure 7). Recombinants are selected by PCR using a primer based on the neo gene and a primer homologous to the 5 'sequences of mouse of the KpnI site in region D.

Alternatively, the heavy chain locus it is canceled by breaking the coding region of the gene. This strategy involves the same 15 kb KpnI fragment used in the previous strategy. The 1.1 kb insert of pMCIneo is inserted into the only BamHI site of exon II, and the HSV tk gene is added to the 3 'end of KpnI. Then double events are selected crosslinking on both sides of the neo insert, which eliminate the gene tk. These are detected from clone groups selected by PCR amplification. One of the PCR primers derives from neo sequences and the other mouse sequences outside the vector of direction. Functional breakdown of immunoglobulin loci of Mouse is presented in the examples.

       \ newpage
    

Non-Human Transgenic Animals Containing Transgenes from Reorganized Immunoglobulin Heavy and Light Chain

An underlying premise of mice previously described transgenic containing transgenes of non-reorganized Ig minocus is that it is possible to generate a complete antibody repertoire not including all segments Variable genes found in the immunoglobulin locus natural. Theoretically, it is possible to reduce the number of sequences different that contribute to the primary repertoire without reducing the secondary repertoire Provided there is enough diversity in the primary repertoire to trigger a response dependent on T cells for any given antigen, hypermutation somatic must be able to deliver a high antibody affinity against this antigen.

This concept is considered an additional phase in the circumstances described below only for the purpose of reference where a complete repertoire of antibodies is generated heterologous completely by somatic mutation. The site of combination with the antigen is created by the interface between the heavy chain amino terminal domain and amino terminal domain Light chain The remains CDR1, 2 and 3 of each of these domains that interact with the antigen are located in three different loops that connect β chains. As described previously, these regions have the greatest sequence diversity between different antibody molecules that recognize different antigens In this way, the antibody repertoire is determined by sequence diversity in CDR1, 2 and 3. Diversity in CDR1, 2 and 3 that results in a complete antibody repertoire originates from three sources: recombinant diversity, diversity of union and somatic mutation. The recombinant diversity in CDR1 and CDR2 originates from the choice of different V segments which contain different sequences CDR1 and CDR2. The diversity recombinant in CDR3 originates from the choice of different segments D and J. The diversity of union contributes only to the diversity of CDR3, while the somatic mutation, which acts to throughout the entire V region, it contributes to diversity in three determining regions of complementarity. The diversity recombinant and binding together constitute diversity of the primary repertoire (figure 1). In this way, the VDJ union generates a set of primary B cells that express IgM.

Any primary repertoire of B cells that express an IgM molecule on the cell surface with a certain minimal affinity for a foreign antigen, internalize this antigen as IgM and ends the cycle on the cell surface. After the antigen is processed and associated peptides are presented in the cell surface by MHC class II molecules. If it shows up enough foreign antigen on the cell surface, this one triggers a T-cell response that, in turn, triggers the T cell-dependent B cell maturation This is the called secondary response (figure 8). Part of this answer involves hypermutation of the variable portion of the genes of immunoglobulin In this way, a B cell clone that experience a secondary response constantly produces new clones with altered immunoglobulin molecules. Clones with greater affinities for the foreign antigen expand so selective due to auxiliary T cells, leading to maturation of affinity of the expressed antibody. Because the somatic hypermutation takes place along the V region complete, there is no theoretical limitation for the maturation process of affinity.

Given the above, the diversity of CDR1 and 2 is not necessary to generate an antibody response complete. Moreover, the diversity in CDR3, created by the VJ union and VDJ, provides sufficient minimum affinity to trigger the T cell-dependent maturation to produce antibodies from high affinity for a large number of different antigens. This way, methods can be used to generate a wide repertoire of antibodies without primary diversity. This diversity is based on the somatic mutation for the generation of diversity of antibodies During the affinity maturation process, the somatic mutation produces a large number of clones with affinities minors, instead of adults, for the stimulation antigen. The Most of these clones are not selected and die. However, yes one of these clones has an affinity for a new antigen that also is present, this clone expands and experiences a maturation of affinity for the new antigen (figure 9). Consequently, a transgenic mouse with human heavy and light chains reorganized combines to form an antibody that has a low affinity for a known antigen. If this animal is injects the known antigen, its B cells undergo a secondary response that leads to the production of antibodies from high affinity for that antigen. However, if this mouse is first inject a mixture of the known antigen and a new antigen, and then it is subsequently exposed to the new antigen alone, high affinity antibodies are produced against the new antigen by the branching process described above. This strategy has two main advantages: first, the Transgenic constructions are easy to generate; and in second instead, reorganized transgenes are able to exclude allelic and Isotypically reorganization of endogenous mouse genes, doing in this way that it is not necessary to eliminate those genes by homologous recombination as previously described.

The first stage of this strategy is the isolation of heavy and light chain genes reorganized from a human hybridoma that expresses an IgM antibody directed against a known antigen. The ideal hybridoma recognizes an easily available antigen that is capable of generating a good mouse T cell response. There are currently several of these human hybridomas, including several that react with promising antigens such as tetanus toxoid, pseudomonas or gram negative bacteria (reviewed by James and Bourla (1987), J. Immunol. Methods ., 100 , 5-40 ). The complete reorganized heavy chain gene is isolated on a single piece of DNA (approximately 20 kb), while the rearranged κ light chain gene, including the 3 'enhancer, is isolated on a second DNA fragment (approximately 20 kb). Each of these fragments is reconstructed from isolated clones of a λ phage library obtained from hybridoma isolated DNA. Two constructions are generated, a heavy chain construction and a light chain construction.

The heavy chain construct (Figure 10) comprises the 20 kb hybridoma fragment, which contains the rearranged IgM gene, linked to a 25 kb fragment containing the human γ3 and γ1 constant regions followed by a 700 fragment bp containing the 3 'heavy chain rat enhancer (Petterson et al . (1990), Nature , 344 , 165-168). The light chain construct consists of the intact 20 kb DNA piece containing the reorganized κ chain and the 3 'enhancer. These two constructs are injected together so that they are integrated into a single mouse genome site. Transgen mRNA expression is assayed in transgenic mice by Northern blot analysis. FACS analysis is then performed on samples of blood from the tail of the mouse to detect the surface expression of the protein cells encoded by the transgene. The mice are then immunized with the antigen recognized by the original hybridoma. ELISA and FACS analyzes with tail blood are performed to detect the class change. Finally, the ability of mice to respond to several different antigens is tested by co-injection of a panel of antigens in conjunction with the original antigen. Tail blood is analyzed by ELISA to detect the production of high affinity human IgG antibodies directed against individual antigens.

To use this transgenic mouse to generate human antibodies directed against a specific antigen, preferably that antigen is first injected together with the antigen associated with the hybridoma from which the genes This hybridoma associated antigen is called a coantigen (sometimes secondary antigen) and the new antigen simply the antigen (or primary antigen). If possible, the antigen secondary chemically intersects with the primary antigen before of the injection. This causes the first antigen to internalize and be presented by the B cells presenting the primary transgene, thus ensuring the existence of a group of T cells activated auxiliaries that recognize the first antigen. A program Typical immunization is as follows. Day 1: they are injected into mice i.p. the first antigen mixed with, or crosslinked with, the second antigen in Freund's complete adjuvant. Day 14: inject the first antigen (without the second antigen) i.p. in adjuvant Freund's incomplete. Day 35: the injection is repeated with the first Freund's incomplete adjuvant antigen. Day 45: rehearsal of the ELISA antibody response in tail blood samples. Day 56: the injection of animals that respond well is repeated with Freund's incomplete adjuvant antigen. Day 59: merger of animal spleens with good response.

Alternatively, the hybridoma recognized antigen from which Ig genes were isolated is used as an immunogen. New transgenic hybridomas are then isolated from the immunized animal that expresses somatically mutated versions of the original antibody. These new antibodies will have a higher affinity for the original antigen. This "remodeling" antibody procedure can also be applied to antibody genes generated by CDR grafting (Pub. PE No. 239400, published September 30, 1987) or isolated from bacterial expression libraries (WD Huse et al. (1989) Science , 246 , 1275) or phage (T. Clackson et al . (1991) Nature , 352 , 624).

Transgenic Non-Human Animals Containing Transgenes of Heavy and / or Light Chain of Immunoglobulin Reorganized or Not Reorganized

Previous embodiments described the use of heavy and light immunoglobulin chain transgenes completely reorganized or completely unorganized to produce non-human transgenic animals capable of producing a heterologous antibody However, the present invention also includes transgenic mice that contain at least one transgene of reorganized immunoglobulin and at least one transgene of non-rearranged immunoglobulin and can be produced using any of the unorganized and reorganized transgenes mentioned above in combination to provide heavy and light chain transgenes to the transgenic mouse. In this aspect, the unorganized transgene comprises a genomic transgene or immunoglobulin heavy chain minilocus and transgene rearranged comprises a light chain transgene of reorganized immunoglobulin.

The combination of the reorganized transgene and the unorganized transgene provides an intermediate level of diversity within the primary repertoire B cells. This way, although the primary diversity in CD1, CD2 and CD3 in the reorganized transgene is fixed in the primary repertoire cell B, the primary diversity in CDR1, CDR2 and CDR3 produced by the reorganization of the unorganized transgene provides a population of primary repertoire of B cells that have a higher potential diversity that the B cell clone obtained when it they use heavy and lightly rearranged chain transgenes. This primary diversity provides expanded secondary diversity when these cells respond to a foreign antigen by means of somatic mutation

Nucleic acids

In nucleic acids, the term "substantial homology" indicates that two nucleic acids, or designated sequences thereof, when optimally aligned and compared, are identical, with appropriate nucleotide insertions or deletions, at least about 80% of the nucleotides, usually in at least about 90 to 95%, and more preferably in at least about 98 to 99.5% of the nucleotides. Alternatively, there is substantial homology when the segments hybridize under conditions of selective hybridization with the complementary chain. Nucleic acids may be present in whole cells, in a cell lysate or in a partially purified or substantially pure form. A nucleic acid is "isolated" or "substantially pure" when purified from other cellular components or other contaminants, for example, other nucleic acids or cellular proteins, by conventional techniques, including alkaline / SDS treatment, CsCl banding, chromatography in column, agarose gel electrophoresis and other methods well known in the art. See , F. Ausubel, et al ., Current Protocols in Molecular Biology , Greene Publishing and Wiley-Interscience, New York (1987).

The nucleic acid compositions of the present invention, although they are often in a native sequence (with the exception of modified and similar restriction sites) from cDNA, genomic DNA or mixtures thereof, may be mutated according to conventional techniques to provide gene sequences In the case of the coding sequences, these mutations can affect the amino acid sequence when it wish. In particular, DNA sequences are contemplated substantially homologous or derived from native sequences V, D, J, constants, change and other sequences described in this document (in which "derivative" indicates that a sequence is identical or modified from another sequence).

A nucleic acid is "bound together operational "when located in a functional relationship with another nucleic acid sequence For example, a promoter or enhancer is operatively linked to a coding sequence if it affects the sequence transcript. Regarding the sequences transcription regulators, operatively linked means that the DNA sequences that are joining are contiguous and, when it is necessary to join two protein coding regions, They are contiguous and in reading phase. In the case of the sequences exchange, operatively linked indicates that the sequences are capable of effecting exchange recombination.

Specific Preferred Embodiments

A preferred embodiment of the invention is a mouse containing a single copy of the transgene described in the example 14 (pHC2) crossed with a mouse containing a single copy of the transgene described in example 16, and the cross offspring with the animal in which JH has been deleted described in the Examples 9 and 12. Animals cross to homozygosis to Each of these three traits. These animals have the following genotype: a single copy (by haploid series of chromosomes) of a non-rearranged human heavy chain minilocus (described in the example 14), a single copy (by haploid series of chromosomes) of a reorganized human light chain construction? (described in example 16), and a deletion in each chain locus Endogenous mouse weighing that eliminates all JH segments functional (described in examples 9 and 12). These animals are cross with mice that are homozygous for the deletion of JH segments (examples 9 and 12) to produce a offspring that is homozygous for JH deletion and hemicigota for constructions Heavy and light chain human. The resulting animals are given inject antigens and are used for antibody production human monoclonal against these antigens.

B cells isolated from said animal are monospecific with respect to human heavy and light chains since they contain only a single copy of each gene. In addition, they will be monospecific with respect to human or heavy chains of mouse, since none of the endogenous copies of the chain genes heavy mouse is functional by virtue of the deletion that encompasses the JH region introduced as described in examples 9 and 12. In addition, a substantial fraction of B cells will be monospecific with respect to human light chains or of mouse, since the expression of the single copy of the chain gene lightweight reorganized human kappa will exclude allelic e Isotypically reorganization of the κ chain genes and endogenous mouse lambda in a significant fraction of cells B.

The transgenic mice of the embodiment preferred will show immunoglobulin production with a repertoire significant, ideally substantially similar to that of a native mouse In this way, for example, when Ig genes endogenous have been inactivated, total immunoglobulin levels range from about 0.1 to 10 mg / ml of serum, preferably 0.5 to 5 mg / ml, ideally at least about 1.0 mg / ml When a transgene has been introduced into the transgenic mouse able to effect a change from IgM to IgG, the serum ratio of the adult mouse between IgG and IgM is preferably of approximately 10: 1. Of course, the relationship between IgG and IgM will be much less in the immature mouse. In general, more than about 10%, preferably 40 to 80% of spleen B cells and of Lymph node exclusively express human IgG protein.

Ideally, the repertoire will approximate that shown by a non-transgenic mouse, usually it will be at least about 10% higher, preferably 25 to 50% or more. Generally, at least about a thousand different immunoglobulins (ideally IgG) will be produced, preferably from 10 4 to 10 6 or more, depending mainly on the number of different V, J and D regions introduced into the mouse genome These immunoglobulins will typically recognize about half or more of the highly antigenic proteins, including, but not limited to: pigeon cytochrome C, chicken lysozyme, phytolac mitogen, bovine serum albumin, Californian barnacle hemocyanin, hemagglutinin virus influenza, Staphylococcus protein A, whale sperm myoglobin, influenza virus neuraminidase and lambda repressor protein. Some of the immunoglobulins will show an affinity for preselected antigens of at least about 10-7 M -1, preferably from 10-8 M -1 to 10-9 M ^ {-1} or greater.

Although the above describes an embodiment preferred of the transgenic mouse of the invention, other are defined realizations by the description of this document and, more particularly, for the transgenes described in the examples. The The present invention encompasses two of the following categories (I and II) of transgenic mice, while the other two categories Additional (III and IV) are provided for reference only.

I. Transgenic mice that contain a transgene of non-rearranged and light chain heavy chain immunoglobulin reorganized

II. Transgenic mice that contain a transgender of unorganized heavy chain immunoglobulin and light chain not reorganized.

III. Transgenic mice that contain a reorganized heavy chain and chain immunoglobulin transgene lightweight not rearranged, and

IV. Transgenic mice that contain transgenes of reorganized heavy chain and light chain immunoglobulin reorganized

Of these categories of transgenic animal, the order of preference with respect to the invention is as follows
I> II.

Within each of these animal categories transgenic, several possible combinations are anticipated. These combinations include the following:

Category I

(a) Animal of examples 1 and 2 or 19 and 20 crossed with animal of example 7 or 16.

(b) Fragment of injected example 1 or 19 together with the fragment of example 7 or 16.

(c) Animal of example 5 (H, I or J) or 14, 17 or 21 crossed with animal of example 7 or 16.

(d) Construction of example 5 (H) or 14 co-injected with the construction of example 7 or 16.

(e) All of the above crossed with the animal of example 9 or 11, or 12 or 13. They are particularly embodiments preferred all the above crosses with animals of example 9, 12 or 13

Category II

(a) Animal of example 1, 2, 19 or 20 crossed with the animal of example 6, 3, 4, 16, 22 or 23.

(b) Fragment of injected example 1 or 19 together with the fragment of example 2 or 20.

(c) Animals of examples 5 (H, I or J) or 14, 17 or 21 crossed with the animal of example 6 (B, C or D) or 16.

(d) Construction 5 (H) or 14 injected in conjunction with construction 6 (B) or 16.

(e) Animal of example 1, 2, 19 or 20 crossed with the animal of example 6 (B, C or D) or 16.

(f) Animal of example 3, 4, 22 or 23 crossed with the animal of example 5 (H, I or J) or 14, 17 or 21.

(g) All of the above crossed with the animal from example 9, 10, 11, 12 or 13.

Category III

(a) Animal of example 3, 4, 22 or 23 crossed with the animal of example 8 or 15.

(b) Fragment 3 or 23 injected together with the fragment of example 8 or 15.

(c) Animal of example 6 (B, C or D) or 16 crossed with the animal of example 8 or 15.

(d) Construction of example 6 (B) or injected 15 together with the construction of example 8 or 15.

(e) All of the above crossed with the animal from example 9 to 13.

Category IV

(a) Animal of example 7 or 16 crossed with the animal of example 8 or 15.

(b) Construction of injected example 7 or 16 together with the construction of example 8 or 15.

(c) All of the above crossed with the animal from example 9 to 13.

What comes next is presented by way of example and should not be considered a limitation of the scope of claims.

Methods and materials

Transgenic mice are obtained according to Hogan, et al ., "Manipulating the Mouse Embryo: A Laboratory Manual", Cold Spring Harbor Laboratory.

Embryonic stem cells are manipulated according to published procedures (Teratocarcinomas and embryonic stem cells: a practical approach, EJ Robertson, ed. IRL Press, Washington, DC, 1987; Zjilstra, et al ., (1989), Nature , 342 , 435-438; and Schwartzberg, P., et al . (1989), Science , 246 , 799-803).

DNA cloning procedures are performed according to J. Sambrook, et al . in Molecular Cloning: A Laboratory Manual, 2nd Ed., 1989, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY

The oligonucleotides are synthesized in a oligonucleotide synthesizer from Applied Bio Systems according with the specifications provided by the manufacturer.

The hybridoma cells and antibodies are handled according to "Antibodies: A Laboratory Manual", Ed Harlow and David Lane, Cold Spring Harbor Laboratory (1988).

Example 1 Genomic transgene of human Ig heavy chain

This example describes cloning and microinjection of a heavy chain genomic transgene of Human immunoglobulin that is microinjected into a murine zygote.

Nuclei are isolated from newly obtained human placenta tissue as described by Marzluff, WF, et al . (1985), "Transcription and Translation. A Practical Approach", BD Hammes and SJ Higgins, eds, p. 89-129, IRL Press, Oxford). Isolated nuclei (or human sperm cells washed with PBS) are included in a low melting agarose matrix and lysed with EDTA and κ proteinase to expose the high molecular weight DNA, which is then digested in the agarose with the NotI restriction enzyme as described by M. Finney in Current Protocols in Molecular Biology (F. Ausubel, et al ., eds. John Wiley & Sons, Suppl. 4, 1988, Section 2.5.1).

The DNA digested with NotI is then fractionated by pulsed field gel electrophoresis as described in Anand, R. et al ., (1989), Nucl. Acids Res ., 17 , 3425-3433. Enriched fractions in the NotI fragment are tested by Southern hybridization to detect one or more of the sequences encoded by this fragment. These sequences include the heavy chain segments D, segments J and constant regions µ and γ1 together with representatives of the 6 VH families (although Berman, et al . (1988), supra identify this fragment as a 670 kb fragment from of HeLa cells, we have discovered that it is an 830 kb fragment from placental DNA and human sperm). Fractions containing this NotI fragment (see Figure 4) are pooled and cloned into the NotI site of the pYACNN vector in yeast cells. Plasmid pYACNN is prepared by digestion of pYAC-4 Neo (Cook, H. et al . (1988), Nucleic Acids Res ., 16 , 11817) with EcoRI and by ligation in the presence of oligonucleotides 5 '-AAT TGC GGC CGC - 3'.

YAC clones containing the heavy chain NotI fragment are isolated as described in Brownstein, et al . (1989), Science , 244 , 1348-1351, and Green, E., et al . (1990), Proc. Natl Acad. Sci. USA , 87 , 1213-1217. The cloned NotI insert is isolated from high molecular weight yeast DNA by pulsed field gel electrophoresis as described in M. Finney, op. cit .. The DNA is condensed by the addition of 1 mM spermine and microinjected directly into the nucleus of single cell embryos described previously.

Example 2 Genomic Ig heavy chain discontinuous transgene

A 110 kb SpeI fragment of genomic DNA human containing VH6, segments D, segments J, the region constant µ and part of the constant region γ (see the Figure 4) is isolated by cloning in YAC as described in the Example 1.

A 570 kb NotI fragment is isolated upstream of the 670-830 kb NotI fragment described above, containing multiple copies of VI through V5, as described (Berman, et al . (1988), supra ). Two 570 kb NotI fragment are detected. Each of them contains multiple segments V).

The two fragments are injected together in the nucleus of a single mouse cell embryo as described in example 1.

Joint injection of two DNA fragments different will normally result in the integration of the two fragments at the same insertion site within the chromosome. By therefore, approximately 50% of transgenic animals resulting that contain at least one copy of each of the two fragments will have the segment V segment fragment inserted above the fragment that contains the constate region. Of these animals, 50% will perform the union of V to DJ by DNA inversion and 50% will do it by deletion, depending on the orientation of the 570 kb NotI fragment in relation to the fragment position 110 kb SpeI. DNA is isolated from animals resulting transgenic and the ability to express is tested human immunoglobulin molecules of animals that, as it has been discovered, they contain the two transgenes by hybridization of Southern blotting (especially animals that contain both multiple human V segments and region genes human constant).

Example 3 Genomic Light Chain Transgen κ of Human Ig Formed by Homologous Recombination In Vivo

In Lorenz, W., et al . (1987), Nucl. Acids Res ., 15 , 9667-9677 a map of the human κ light chain has been described and is depicted in Figure 11.

A fragment (a) of XhoI at NotI of 450 is isolated kb that includes all of the following: C?, the 3 'enhancer, all segments J and at least five different segments V (a) and it is microinjected into the nucleus of single cell embryos as it described in example 1.

Example 4 Genomic Light Chain Transgen κ of Human Ig Formed by Homologous Recombination In Vivo

It is isolated as described in example 1 (see Figure 11) a 750 kb MluI to NotI fragment (b) that includes all of the above plus at least 20 more V segments and digested with BssHII to produce a fragment of approximately 400 kb (c).

The fragment (a) of XhoI at NotI of 450 kb plus the fragment (c) of MluI to BssHII of 400 kb have a sequence overlap defined by restriction sites BssHII and XhoI shown in figure 11. The homologous recombination of these two fragments after microinjection of a mouse zygote gives rise to a transgene that contains at least about 15-20 additional V segments in addition to those found in the fragment 450 kb XhoI / NotI (example 3).

Example 5 Heavy Chain Minilocus Construction A. Construction of pGP1 and pGP2

PBR322 is digested with EcoRI and with StyI and ligated with the following oligonucleotides to generate pGP1 containing an insert of 147 base pairs with restriction sites shown in figure 13. The general overlap of these oligos It is also shown in Figure 13.

The oligonucleotides are:

one

This plasmid contains a large polylinker flanked by rare NotI cutting sites to construct large inserts that can be isolated from vector sequences for microinjection. The plasmid is based on pBR322 which has a relatively low number of copies compared to pUC-based plasmids (pGP1 retains the copy number control region of pBR322 near the origin of replication). A low number of copies reduces the potential toxicity of the inserted sequences. In addition, pGP1 contains a strong transcription termination sequence derived from trpA (Christie, GE, et al . (1981), Proc. Natl. Aca. Sci. USA ) inserted between the ampicillin resistance gene and the polylinker. This also reduces the toxicity associated with certain inserts by preventing transcription transcription initiated from the ampicillin promoters.

Plasmid pGP2 derives from pGP1 by introducing a additional restriction site (SfiI) in the polylinker. pGP1 se digested with MluI and SpeI to cut recognition sequences in the polylinker portion of the plasmid.

The digested pGP1 in this way is linked following adapter oligopeptides to form pGP2.

2

pGP2 is identical to pGP1 except because it contains an additional Sfi I site located between the MluI and SpeI sites. This allows the inserts to be completely cleaved with SfiI so As with NotI.

B. Construction of pRE3 (3 'rat enhancer)

Heavy chain constructions include an enhancer sequence located downstream of the region rat constant

The 3 'enhancer of the heavy chain region described by S. Pettersson, et al ., (1990), Nature , 344 , 165-168) is isolated and cloned. The 3 'enhancer sequence of rat IGH is amplified by PCR using the following oligonucleotides:

3

The double stranded DNA formed in this way that encodes the 3 'enhancer is cut with BamHI and SphI and cloned into pGP2 cut with BamHI / SphI to produce pRE3 (3 'enhancer of rat).

C. Cloning of the J- \ region human

A substantial portion of this region is cloned combining two or more isolated fragments from inserts of the lambda phage. See figure 14.

A 6.3 kb BamHI / HindIII fragment that includes all human J segments (Matsuda, et al ., (1988), EMBO J. , 7 , 1047-1051; Ravetech, et al . (1981), Cell is isolated. , 27 , 583-591) from a human genomic DNA library using the oligonucleotide GGA CTG TGT CCC TGT GTG ATG CTT TTG ATG TCT GGG GCC AAG.

Similarly, an adjacent 10 kb HindIII / BamII fragment containing the enhancer, and exons encoding the exchange region and the constant region (Yasui, et al . (1989), Eur. J. Immunol ., Is isolated. 19 , 1399-1403) using the oligonucleotide:

CAC CAA GTT GAC CTG CCT GGT CAC AGA CCT GAC CAC CTA TGA

Similarly, a 3 'BamHI fragment is isolated adjacent 1.5 kb using the pMUM clone insert as a probe (pMUM it is a 4 kb EcoRI / HindIII fragment isolated from a Human genomic DNA library with the oligonucleotide:

4

exon 1 of mu membrane) and is cloned in pUC19.

pGP1 is digested with BamHI and with BglII followed by Calf intestinal alkaline phosphatase treatment.

Fragments (a) and (b) of Figure 14 are they clone in the digested pGP1. Then a clone is isolated that is oriented so that the BamHI 5 'site is destroyed by fusion BamHI / Bgl. It is identified as pMU (see Figure 15). pMU se digested with BamHI and the fragment (c) of Figure 14 is inserted. Orientation is checked by digestion with HindIII. Plasmid resulting pHIG1 (figure 15) contains an 18 kb insert that encodes segments J and C.

D. Cloning of the C region

pGP1 is digested with BamHI and with HindIII followed of calf intestinal alkaline phosphatase treatment (figure 14). Fragment (b) of Figure 14 treated in this manner and the fragment (c) of Figure 14 are cloned into pGP1 cut with BamHI / HindIII. The appropriate orientation of the fragment (c) is check by digestion with HindIII to form pCON1 containing a 12 kb insert encoding the C region.

Whereas pHIG1 contains J segments, change sequences and µ sequences in its 18 kb insert with a 3 'SfiI site and a 5' SpeI site in a polylinker flanked by NotI sites, it will be used for reorganized VDJ segments. pCON1 is identical except that it lacks a J region and contains only a 12 kb insert. The use of pCON1 in the construction of a fragment containing rearranged VDJ segments will be described below.
Nice

E. Constant Region Cloning γ-1 (pREG2)

The cloning of the human region γ-1 is depicted in Figure 16.

Yamamura, et al . (1986), Proc. Natl Acad. Sci. USA , 83 , 2152-2156 reported on the expression of membrane-bound human γ-1 from a transgenic construct that had been partially deleted in the integration. Their results indicate that the 3 'BamHI site delimits a sequence that includes the reorganized and changed transmembrane copy of the gamma gene with a VC intron of less than 5 kb. Therefore, in the unreorganized or changed gene, the entire change region is included in a sequence that begins less than 5 kb from the 5 'end of the first constant exon of γ-1. Therefore, it is included in the 5.0 kb 5 'HindIII fragment (Ellison, JW, et al . (1982), Nucleic Acids Res ., 10 , 4071-4079). Takahashi, et al ., (1982), Cell , 29 , 671-679 also report that this fragment contains the change sequence, and this fragment together with the 7.7 kb HindIII to BamHI fragment must include all the sequences that We need for transgenic construction.

Phage clones are identified that contain the γ-1 region and are isolated using the following oligonucleotide that is specific for the third exon of γ-1 (CH3).

       \ vskip1.000000 \ baselineskip
    

5

       \ vskip1.000000 \ baselineskip
    

A 7.7 kb HindIII to BglII fragment (fragment (a) of Figure 16) is cloned into pRE3 cut with HIndIII / BglII to form pREG1. The HindIII fragment upstream 5.3 kb (fragment (b) of Figure 16) is cloned into digested pREG1 with HindIII to form pREG2. The correct orientation is confirmed by digestion with BamHI / SpeI.

F. Combination of Cγ and C

The pHIG1 plasmid described previously contains human J segments and exons of the constant region C. For provide a transgene that contains the gene segments of constant region C, pHIG1 was digested with SfiI (Figure 15). He plasmid pREG2 was also digested with SfiI to produce an insert 13.5 kb containing exons from the human Cγ region and the 3 'rat enhancer sequence. These sequences were combined to produce the plasmid pHIG3 '(figure 17) containing the human J segments, the constant human C region, the region human Cγ1 constant and the 3 'rat enhancer contained in a 31.5 kb insert.

A second plasmid is constructed that encodes C µ human and C γ human without the J segments digesting pCON1 with SfiI and combining it with the SfiI fragment that contains the human Cγ region and the 3 'rat enhancer by digestion of pREG2 with SfiI. The resulting plasmid, pCON (Figure 17) contains a 26 kb NotI / SpeI insert containing human C, γ1 human and the 3 'rat enhancer sequence.

G. Cloning of segment D

The strategy for cloning human D segments is depicted in Figure 18. Phage clones of the human genomic library containing the D segments are identified and isolated using specific probes for sequences from the region of diversity (Y. Ichihara, et al. ., (1988), EMBO J. , 7 , 4141-4150). The following oligonucleotides are used:

6

A 5.2 kb XhoI fragment (fragment (b) of the Figure 18) containing DLR1, DXP1, DXP'1 and DA1, it is isolated from of a phage clone identified with oligo DXP1.

A 3.2 kb XbaI fragment (fragment (c) in the Figure 18) which contains DXP4, DA4 and DK4 is isolated from a clone of phage identified with oligo DXP4.

Fragments (b), (c) and (d) of Figure 18 are combine and clone at the XbaI / XhoI site of pGP1 to form pHIG2 which contains a 10.6 kb insert.

This cloning is done sequentially. In first, the 5.2 kb fragment (b) of Figure 18 and the 2.2 kb fragment (d) of Figure 18 are treated with phosphatase calf intestinal alkaline and cloned in pGP1 digested with XhoI and with XbaI. The resulting clones are selected with the inserts 5.2 and 2.2 kb. Half of the clones with a positive result in the tests with the 5.2 and 2.2 kb inserts have the 5.2 insert kb in the proper orientation as determined by digestion with BamHI Then, the 3.2 kb XbaI fragment of Figure 18 is cloned in this intermediate plasmid containing fragments (b) and (d) to form pHIG2 (figure 9). This plasmid contains segments of diversity cloned in the polylinker with a unique SfiI 5 'site and a unique site SpeI 3 '. The complete polylinker is flanked by NotI sites.

H. Heavy Chain Minilocus Construction

The following describes the construction of a human heavy chain minilocus containing one or more segments V.

An unorganized segment V corresponding to the one identified as segment V contained in the hybridoma of Newkirk, et al . (1988), J. Clin. Invest ., 81 , 1511-1518, using the following oligonucleotide:

7

       \ vskip1.000000 \ baselineskip
    

A segment restriction map is determined V not reorganized to identify unique restriction sites which provide after digestion a DNA fragment that has a approximate length of 2 kb containing segment V not rearranged together with the 5 'and 3' flanking sequences. The 5 'primer sequences will include promoters and other sequences regulators, while the 3 'flanking sequence provides recombination sequences necessary for binding V-DJ This segment V insert of approximately 3.0 kb is cloned in the polylinker of pGB2 to form pVH1.

pVH1 is digested with SfiI and the fragment The resulting is cloned into the SfiI site of pHIG2 to form pHIG5 '. Since pHIG2 contains only D segments, the resulting plasmid pHIG5 'contains a single segment V together with segments D. The size of the insert contained in pHIG5 is 10.6 kb plus the size of the V segment insert.

The pHIG5 insert is cleaved by digestion with NtoI and SpeI and is isolated. pHIG3 ', which contains segments J, Cµ and Cγ1, is digested with SpeI and NotI and the 3 kb fragment containing ... kb containing these sequences and the 3' rat enhancer sequence is isolated. These two fragments are combined and ligated into pGP1 digested with NotI to produce pHIG containing the insert encoding a V segment, nine D segments, six functional J segments, C µ, C γ and the 3 'rat enhancer. The size of this insert is approximately 43 kb plus the insert size of the
segment V.

I. Heavy Chain Minilocus Construction by Homologous Recombination

As indicated in the previous section, the pHIG insert is approximately 43 to 45 kb when a single V segment is used. That insert size is at the limit or near the limit of those that can be easily cloned into plasmid vectors. To provide the use of a greater number of V segments then homologous recombination is described in vivo of overlapping DNA fragments which upon homologous recombination within a zygote or an ES cell form a transgene containing the enhancer sequence 3 'rat , the human C, the human C, the human J segments, human D segments and a multiplicity of human V segments.

A 6.3 kb BamHI / HindIII fragment containing the human J segments (see fragment (a) of Figure 14) are clone in pHIG5 'digested with MluI / SpeI using the following adapters:

8

The result is a plasmid called pHIG5'O (overlap). The insert contained in this plasmid contains human V, D and J segments. When the single V segment of pVH1 is used, the size of this insert is approximately 17 kb plus 2 kb. This insert is isolated and combined with the pHIG3 'insert containing the human sequences J, C µ, γ1 and the 3' rat enhancer. Both inserts contain human J segments that provide approximately 6.3 kb of overlap between the two DNA fragments. When injected together in the mouse zygote, homologous recombination takes place in vivo that generates a transgene equivalent to the insert contained in pHIG.

This strategy provides the addition of a multiplicity of V segments within the transgene formed in vivo . For example, instead of incorporating a single V segment into pHIG5 ', a multiplicity of V segments contained in (1) isolated genomic DNA, (2) linked DNA derived from genomic DNA, or (3) DNA encoding a repertoire are cloned of synthetic V segments; within pHIG2 at the SfiI site to generate pHIG5 'V N. Then, fragment (a) of segments J of Figure 14 is cloned in pHIG5 'V N and the insert is isolated. This insert now contains a multiplicity of segments V and segments J that overlap with segments J contained in the insert isolated from pHIG3 '. When introduced together in the nucleus of a mouse zygote, homologous recombination takes place to generate in vivo the transgene encoding multiple V segments and multiple J segments, multiple D segments, the C region, the C γ1 region ( all of human origin) and the 3 'rat enhancer sequence.

J. Construction of a Heavy Chain Minilocus by Joint Fragment Injection of the Synthetic VH Region In conjunction with a Heavy Chain DJC Construction

Synthetic fragments of the region are generated V_ {H} and are isolated as previously described. These fragments are injected together with the purified NotI insert of the pHIG plasmid (or a version of pHIG that does not contain any segment V). The DNA fragments injected together are inserted into a single site on the chromosome. Some of the animals resulting transgenics will contain transgenic inserts that have synthetic V regions located in adjacent position and chain above the sequences of the pHIG construct. Those animals they will have a primary repertoire of human heavy chain greater than animals described in example 5 (H).

Example 6 Light Chain Minilocus Construction A. Construction of pE µ1

The construction of pE µ1 is shown in Figure 21. The mouse heavy chain enhancer is isolated from the 678 bp XRI fragment to EcoRI (J. Banerji, et al . (1983), Cell , 33 , 729- 740) from phage clones using oligo:

       \ vskip1.000000 \ baselineskip
    

9

       \ vskip1.000000 \ baselineskip
    

This Eµ fragment is cloned into digested pGP1 with EcoRV / XbaI filling the blunt ends at the EcoRI site. He The resulting plasmid is called pEmu1.

B. Construction of the Light Chain Minilocus \kappa

The κ construction contains at least one human V \ kappa segment, the five human J \ kappa segments, the human JC \ kappa enhancer, the exon from the human κ constant region and, ideally, the 3 '\ kappa enhancer human (K. Meyer, et al . (1989), EMBO J. , 8 , 1959-1964). The κ enhancer in the mouse is 9 kb downstream of C κ. However, it has not yet been identified in humans. In addition, the construction contains a copy of the JC µ enhancers of the mouse heavy chain.

The minilocus is built from four component fragments:

(a) a 16 kb SmaI fragment containing the exon C \ kappa human and the human enhancer 3 'by analogy with the mouse locus (fragment (a) in figure 20);

(b) an adjacent 5 kb SmaI fragment of 5 kb, which it contains the five segments J (fragment (b) in Figure 20);

(c) the heavy chain intronic enhancer of mouse isolated from pE µ1 (this sequence is included for induce light chain construction expression as soon as possible possible in the development of B cells. As the genes of the heavy chain are transcribed before chain genes lightweight, this heavy chain enhancer presumably is active at an earlier stage than the intronic enhancer κ); Y

(d) a fragment that contains one or more segments V.

The preparation of this construction is like follow. Human placenta DNA is digested with SmaI and fractionated into agarose gel by electrophoresis. Similarly, DNA is digested of human placenta with BamHI and is fractionated by electrophoresis. The 16 kb fraction is isolated from the gel digested with SmaI and the 11 kb region is similarly isolated from the gel that contains the DNA digested with BamHI.

The 16 kb SmaI fraction is cloned in lambda FIX II (Stratagene, La Jolla, California) that has been digested with XhoI, and it is treated with the klenow fragment of DNA polymerase to fill the product of restriction digestion of XhoI. The linkage of the 16 kb SmaI fraction destroys the SmaI sites and leaves intact XhoI sites.

The 11 kb BamHI fraction is cloned into λ EMBL3 (Stratagene, La Jolla, California) that is digested with BamHI before cloning.

The clones of each library were probed with the C \ kappa specific oligo:

10

a 16 kb XhoI insert that was subcloned into pE µ1 cut with XhoI, so that C \ kappa is adjacent to the SmaI site. The resulting plasmid was named pKap1. See the figure 22.

The C \ specific oligonucleotide above is used to probe the λ EMBL3 / BamHI library for identify an 11 kb clone corresponding to fragment (d) of the Figure 20. A 5 kb SmaI fragment (fragment (b) of Figure 20) it is subcloned and subsequently inserted into pKap1 digested with SmaI. Plasmids containing the correct orientation of the segments J, C? And the enhancer E? Are called pKap2.

After this, one or more segments are subcloned V? At the MluI site of pKap2 to produce plasmid pKapH which encodes the human V \ kappa segments, the segments Human JK, the human CK segments and the enhancer E \ human. This insert is cleaved by digestion of pKapH with NotI and purified by agarose gel electrophoresis. The insert purified in this way it is microinjected into the pronucleus of a Mouse zygote as previously described.

C. Construction of Light Chain Minilocus \ kappa by Homologous Recombination In Vivo

The 11 kb BamHI fragment (fragment (d) of the Figure 20) is cloned into pGP1 digested with BamHI so that the 3 'end is towards the SfiI site. The resulting plasmid is called pKAPint. One or more V \ kappa segments are inserted into the polylinker between the BamHI and SpeI sites in pKAPint to form pKapHV. The pKapHV insert is cleaved by digestion with NotI and is purify The pKap2 insert is cleaved by digestion with NotI and is purify Each of these fragments contains regions of homology, since the pKapHV fragment contains a sequence of 5 kb of DNA that includes the J? segments that is substantially homologous to the 5 kb SmaI fragment contained in the insert obtained from pKap2. In this way, these inserts are capable of recombining homologously when microinjected into a mouse zygote to form a transgene encoding V? J \ kappa and C \ kappa.

D. Construction of the Light Chain Minilocus \ kappa by Joint Injection of Synthetic V \ kappa Region Fragments Together with JC Light Chain Constructions

Synthetic fragments of the region are generated V? And are isolated as previously described. These DNA fragments are injected together with the NotI insert purified from plasmid pKap2 or plasmid pKapH. Fragments of DNA injected together are inserted into a single site of the chromosome. Some of the resulting transgenics will contain transgene inserts that have localized synthetic V regions adjacent and upstream of the sequences in the construction pKap2 or pKapH. These animals will have a primary repertoire of human κ light chain greater than those described in the example 6 (B).

Example 7 Isolation of Genomic Clones Corresponding to Copies Reorganized and Expressed of Light Chain Genes \ kappa de Immunoglobulin

This example describes the cloning of genes from κ immunoglobulin light chain from cultured cells that express an immunoglobulin of interest. These cells can contain multiple alleles of a particular immunoglobulin gene. For example, a hybridoma could contain four copies of the gene from light chain \ kappa, two copies of the companion cell line of fusion and two copies of the original B cell expressing the immunoglobulin of interest. Of these four copies, only one encodes the immunoglobulin of interest, despite the fact that Several of them may have been reorganized. The procedure described in this example allows selective cloning of the copy expressed from light chain κ.

A. Double Chain cDNA

Human hybridoma cells, or of the lymphoma, or another cell line that synthesizes the form of cell surface or secreted or both of IgM with a light chain κ for the isolation of poly A + RNA. Next, the RNA is used for the synthesis of cDNA primed with oligo dT using the enzyme reverse transcriptase. Then, the single stranded cDNA is isolated and add G residues to the 3 'end using the polynucleotide enzyme terminal transferase The single stranded cDNA is then purified with G tail and is used as a template for the synthesis of the second chain (catalyzed by the enzyme DNA polymerase) using the following oligonucleotide as primer:

eleven

The double stranded cDNA is isolated and used to determine the nucleotide sequence of the 5 'end of mRNAs encoding the light and heavy chains of the molecule of expressed immunoglobulin. Then the genomic clones of these genes expressed. The procedure to clone the gene from expressed light chain is described in part B presented to continuation.

B. Light Chain

The double stranded cDNA described in part A is denatures and is used as a template for a third cycle of DNA synthesis using the following oligonucleotide primer:

12

This primer contains specific sequences for the constant portion of the light chain messenger \ kappa (TCA TCA GAT GGC GGG AAG ATG AAG ACA GAT GGT GCA) as well as sequences unique ones that can be used as a primer for PCR amplification of the newly synthesized DNA chain (GTA CGC CAT ATC AGC TGG ATG AAG). The sequence is amplified by PCR using the two primers. following oligonucleotides:

13

The amplified sequence is then purified by Gel electrophoresis PCR and is used as a template for dideoxy sequencing reactions using the following oligonucleotide as primer:

5 '- GAG GTA CAC TGA CAT ACT GGC ATG - 3'

Then the first 42 nucleotides of the sequence to synthesize a single probe for the isolation of gene from which the messenger of the immunoglobulin This 42 nucleotide synthetic DNA segment is will name from now o-kappa.

Next, a Southern blot of DNA isolated from the cell line that expresses the immunoglobulin and digested individually and in combinations of couples with several different restriction endonucleases, including SmaI, is probed with the oligonucleotide single o-kappa labeled with 32 P. It identifies a single site for restriction endonuclease upstream of the segment V reorganized.

Subsequently, the DNA of the cell line that expresses the Ig is cut with SmaI and with a second enzyme (BamHI or KpnI if there is a SmaI site within segment V). Any extreme resulting non-blunt is treated with the enzyme T4 DNA polymerase for give blunt ends to DNA molecules. Then they are added linkers encoding restriction sites (BamHI, EcoRI or XhoI depending on which site does not exist in the fragment) and are cut with the corresponding linker enzyme to give DNA fragments with BamHI, EcoRI or XhoI ends. Then the DNA is fractionated by size in agarose gel electrophoresis and the fraction that includes the DNA fragment covering the expressed V segments is cloned into lambda EMBL3 or lambda FIX (Stratagene, La Jolla, California). The clones containing segment V are isolated using the single probe o-kappa DNA is isolated from clones positive and subcloned into the polylinker of pKap1. The clone The result is called pRKL.

Example 8 Isolation of Genomic Clones Corresponding to Copies Reorganized Expresses of Heavy Chain Genes \ mu Immunoglobulin

This example describes the cloning of genes heavy chain immunoglobulin from cultured cells that express an immunoglobulin of interest. The procedure described in this example allows selective cloning of the copy expressed from a heavy chain gene.

Double stranded cDNA is prepared and isolated as is described in part A of example 7. The double stranded cDNA is denatures and is used as a template for a third cycle of DNA synthesis using the following oligonucleotide primer:

14

This primer contains specific sequences for the constant portion of the heavy chain messenger µ (ACA GGA GAC GAG GGG GAA AAG GGT TGG GGC GGA TGC) as well as unique sequences which can be used as a primer for PCR amplification of the DNA string of new synthesis (GTA CGC CAT ATC AGC TGG ATG AAG). The sequence is amplified by PCR using the two primers. following oligonucleotides:

fifteen

The amplified sequence is then purified by Gel electrophoresis PCR and is used as a template for dideoxy sequencing reactions using the following oligonucleotide as primer:

5 '- GAG GTA CAC TGA CAT ACT GGC ATG - 3'

Then the first 24 nucleotides of the sequence to synthesize a single probe for the isolation of gene from which the messenger of the immunoglobulin This segment of 42 synthetic DNA nucleotides is will name from now o-mu.

Then, a Southern DNA transfer isolated from the cell line expressing Ig and digested individually and in combinations of couples with several different restriction endonucleases including MluI (MluI is a rare cutting enzyme that cleaves between segment J and mu CH1), is probed with the single oligonucleotide o-mu marked with 32 P. A single restriction site is identified for endonuclease upstream of the rearranged V segment.

Then, the DNA from the cell line which expresses Ig is cut with MluI and a second enzyme. Then the MluI or SpeI adapter linkers are linked at the ends and are cut to turn the site upstream into a MluI site or SpeI. Then, the DNA is divided by size by electrophoresis in agarose gel and the fraction that includes the DNA fragment that encompasses the segment V expressed is cloned directly into the plasmid pGPI. Clones containing segment V are isolated using the single o-mu probe, and the insert is subcloned into the plasmid pCON2 cut with MluI or with MluI / SpeI. Plasmid The result is called pRMGH.

Example 9 Deletion of the Mouse Heavy Chain Gene by Recombination Homologous

This example describes the deletion of the endogenous mouse heavy chain gene by homologous recombination in embryonic stem cells (ES) (Zjilstra, et al . (1989), Nature , 342 , 435-438) followed by transplantation of those ES cells into a embryo in a mouse blastocyst state so that ES cells colonic the germ line of the resulting chimeric mouse (Teratocarcinomas and embryonic stem cells: a practical approach, EJ Robertson, ed., IRL press, Washington, DC, 1987).

The construction of a DNA sequence that will recombine homologously on the mouse chromosome so that the heavy chain segments J be deleted, being removed from this way the possibility of a satisfactory gene reorganization in the locus of the heavy chain. The design of this construction is described below.

Plasmid pGP1 is digested with endonucleases BamHI and BglII restriction and the resulting fragments are religed to form plasmid pGP1d1. Then, this plasmid is used to build the so-called knockout gene construction.

To obtain sequences homologous to the region desired target of the mouse genome, genomic clones of mouse from a phage library derived from non-tissue lymphoid (such as liver) using the oligonucleotide probe J_ {H} specific:

16

A fragment of KpnI is isolated to 3.5 kb EcoRI which hybridizes with this probe from the DNA derived from clones of positive phage. This fragment is subcloned into pGP1d1 digested with KpnI / EcoRI to form plasmid pMKO1.

The neomycin (Neo) and thymidine kinase (TK) resistance genes of the herpes simplex virus are then isolated for recombinant drug selection (M. Capecchi (1989), Science , 244 , 1288-1292) as follows. Plasmid pGEM7 (KJ1) (MA Rudnicki, 3/15/89) is digested with HindIII and the ends become blunt with the klenow form of DNA polymerase I. Then, the DNA is cut with EcoRI and the pGKneo fragment is Isolate and clone into pMKO1 cut with SphI / NaeI using the following oligonucleotide as an adapter:

5 '- AATTGTAC - 3'

The resulting plasmid is called pMKO2. This plasmid contains the neomycin resistance gene flanked by sequences flanking the mouse segments JH. This plasmid can only be used for deletion of the heavy chain gene. Alternatively, the herpes TK gene can be added to the construct to improve the frequency of homologous recombination events in Neo-resistant clones (M. Capecchi (1989), Science , 244 , 1288-1292). This is done as follows. The EcoK PGKTK fragment is isolated to HindIII from plasmid pGEM7 (TK) (MA Rudnicki) and cloned into the KpnI site of pMKO2 using the following oligonucleotides as adapters:

17

The resulting plasmid is called pMKO3.

To further improve overall efficiency of homologous recombination, a construction is then added to the construction large segment of DNA that is homologous to the target sequence. A 13 kb EcoRI fragment that hybridizes with the oligonucleotide specific for C µ described below:

18

This 12 kb fragment includes exons that encode C, or a substantial portion of this fragment that includes the 5 'EcoRI end, isolated from a library of Mouse genomic phage and subcloned into the EcoRI site of pMKO3. He The resulting plasmid is called pMKO4.

The pMKO4 insert is isolated by digestion with NotI and electroporated into ES cells. Homologous recombinant clones used to generate a JH-deleted mouse are isolated as described in Zjilstra, et al . (1989), Nature , 342 , 435-438.

Example 10 Deletion of the Mouse Light Chain Gene by Recombination Homologous

This example describes the deletion of the gene from endogenous mouse light chain by homologous recombination in embryonic stem cells (see previous example).

A recombinant DNA sequence is constructed homologously within the mouse chromosome to delete the exon of the light chain constant region κ. The design of This construction is described below.

A thymidine kinase fragment from BamHII to EcoRI of 2 kb obtained from pGEM7 (TK) Sal (M. A. Rudnicki, Whitehead Institute) and is subcloned into pGP1 digested with BamHI / SfiI using the following oligonucleotide adapter:

5 '- AATTTTG - 3'

The resulting plasmid is called pKKO1.

To obtain sequences homologous to the region desired target of the mouse genome, genomic clones of mouse from a phage library derived from non-tissue lymphoid (such as liver) using chain specific oligo lightweight \ kappa of mouse named o-MKC presented then:

19

DNA is isolated from a positive clone and a 2.3 kb BglII fragment (PS Neumaier and HG Zachau (1983), Nucl. Acids. Res ., 11 , 3631-3656) that hybridizes with the probe is isolated. MK3 The sequence of the o-MK3 probe is as follows:

twenty

This 2.3 kb BglII fragment is subcloned into pKKO1 digested with BamHI so that the 3 'end of the fragment is adjacent to the SfiI site of the polylinker. Plasmid The result is called pKKO2.

The 4 kb SphI to Hpa I DNA fragment that hybridized with the oligonucleotide o-MKC is isolated to from a positive phage clone and subcloned into the plasmid pKKO2 digested with EcoRV and SPhI. The resulting plasmid is called pKKO3.

A 2 kb SalI fragment is isolated to EcoRI of pGEM7 (KJ1) Salt (M. A. Rudnicki, 3/15/89) and is cloned into the BssHII site of plasmid pKKO3 using linker adapters. This is done by first linking a mixture of the three following oligonucleotides with the SalI fragment to EcoRI of 2 kb:

twenty-one

Then, the ligation mixture is digested with BssHII enzyme and binds to plasmid pKKO3 digested with BssHII. He The resulting plasmid is called pKKO4.

The pKKO4 insert is isolated by digesting with NotI and electroporated into ES cells. Homologous recombination clones are isolated and used to generate a mouse in which C? Has been deleted as described in Zjilstra, et al . (1989), Nature , 342 , 435-438.

Example 11 Inactivation of the Kappa Mouse Light Chain Gene by Homologous Recombination

This example describes locus inactivation. endogenous mouse kappa by homologous recombination in stem cells embryonic (ES) followed by the introduction of the mutated gene in the mouse germ line by injection of target ES cells that carry an inactive kappa allele in early-stage mouse embryos (blastocysts).

The strategy is to delete J \ kappa and C \ kappa by homologous recombination with a vector containing sequences of DNA homologous to the mouse kappa locus in which a 4.5 kb segment of the locus, which encompasses the J \ kappa gene and the C \ kappa segments, and is replaced by the selectable marker neo.

Construction of the kappa address vector

Plasmid pGEM7 (KJ1) (MA Rudnicki, Whitehead Institute) contains the neomycin resistance (neo) gene, used for drug selection of transfected ES cells, under the transcriptional control of the mouse phosphoglycerate kinase promoter (pgk) (XbaI / I / TaqI fragment; Adra, CN et al ., (1987) Gene , 60 , 65-74) in the cloning vector pGEM-72f (+). The plasmid also includes a heterologous polyadenylation site for the neo gene derived from the 3 'region of the mouse pgk gene (PvuII / HindIII fragment; Boer, PH, et al ., (1990) Biochemical Genetics , 28 , 299-308) . This plasmid was used as a starting point for the construction of the kappa direction vector. The first stage was the insertion of sequences homologous to the 3 'locus kappa of the neo expression cassette.

Mouse kappa chain sequences were isolated (Figure 25a) from a genomic phage library derived from Liver DNA using oligonucleotide probes specific for the locus C \ kappa:

5 '- GGC TGA TGC TGC ACC AAC TGT ATC CAT CTT CCC ACC ATC CAG - 3 '

and for the gene segment J \ kappa5:

5'- CTC ACG TTC GGT GCT GGG ACC AAG CTG GAG CTG AAA CGT AAG - 3 '

An 8 kb BglII / SacI fragment was isolated which was extended in 3 'position of the mouse segment C \ kappa from a positive phage clone in two pieces, such as a BglII / SacI fragment 1.2 kb and a 6.8 kb SacI fragment, and subcloned into pGEM7 (KJ1) digested with BglII / SacI to generate the plasmid pNEO-K3 '(figure 25b).

A 1.2 EcoRI / SphI fragment was also isolated kb that extended in position 5 'of the region J \ kappa from a positive phage clone. An SphI / XbaI / BglII / EcoRI adapter was linked at the SphI site of this fragment and the resulting EcoRI fragment bound in pNEO-K3 'digested with EcoRI, in it 5 'to 3' orientation than the neo gene and 3 'kappa sequences string down, to generate pNEO-K5'3 '(figure 25c).

The thymidine kinase (TK) gene of herpes simplex virus (HSV) was then included in the construct to allow enrichment of ES clones carrying homologous recombinants, as described in Mansour et al . ((1988) Nature , 336 , 348-352). The HSV TK cassette was obtained from plasmid pGEM7 (TK) (MA Rudnicki), which contains the structural sequences for the HSV TK gene separated by the mouse pgk promoter and polyadenylation sequences as described above for pGEM7 ( KJ1). The EcoRI site of pGEM7 (TK) was modified to obtain a BamHI site and then the TK cassette was cleaved as a BamHI / HindIII fragment and subcloned into pGP1b to generate pGP1b-TK. This plasmid was linearized at the XhoI site and the XhoI fragment of pNEO-K5'3 ', which contained the neo gene flanked by 5'de J \ kappa and 3' C \ kappa genomic sequences, was inserted into pGP1b-TK for generate the address vector J / C KI (figure 25d). The alleged structure of the genomic kappa locus after homologous recombination with J / C K1 is shown in Figure 25e.

Generation and analysis of ES cells with targeted inactivation of a kappa allele

ES AB-1 cells were cultured on layers of mitotically inactive SNL76 / 7 nurse cells (McMahon, AP and Bradley, A. (1990) Cell , 62 , 1073-1085) essentially as described (Robertson, EJ, (1987) in Teratocarcinomas and Embryonic Stem Cells: A Practical Approach . EJ Robertson, ed. (Oxford: IRL Press), p. 71-112).

Kappa chain inactivation J / C K1 vector was digested with NotI and introduced by electroporation into AB-1 cells by the methods described (Hasty, PR, et al . (1991) Nature , 350 , 243-246). The electroporated cells were placed in 100 mm diameter plates at a density of 2-5 x 10 6 cells / plate. After 24 hours, G418 (200 µg / ml active component) and FIAU (0.5 µM) were added to the medium, and drug resistant clones were allowed to develop for 10-11 days. Clones were selected, trypsinized, divided into two portions and further expanded. Half of the cells derived from each clone were then frozen and in the other half homologous recombination between the vector and the target sequences was analyzed.

DNA analysis was performed by Southern blot hybridization. The DNA was isolated from the clones as described (Laird, PW et al ., (1991) Nucl. Acids. Res ., 19 ) by digesting with XbaI and probed with the 800 bp EcoRI / XbaI fragment indicated in the Figure 25e as diagnostic probe. This probe detects a 3.7 kb XbaI fragment in the wild-type locus and a 1.8 kb diagnostic band in a locus that has homologously recombined with the direction vector (see Figures 25a and 25e). Of the 358 clones resistant to G418 and FIAU analyzed by Southern blot analysis, 4 showed the 1.8 kb XbaI band indicative of a homologous recombination in the kappa locus. These 4 clones were further digested with the enzymes BglII, SacI and PstI to verify that the vector was homologously integrated into one of the kappa alleles. When probed with the 800 bp EcoRI / XbaI diagnostic fragment, the BglII, SacI and PstI digestion products of wild-type DNA produced 4.1, 5.4 and 7 kb fragments, respectively, while the presence of a directed kappa allele could be indicated by the 2.4, 7.5 and 5.7 kb fragments, respectively (see Figures 25a and 25e). The 4 positive clones detected by XbaI digestion showed the expected BglII, SacI and PstI restriction fragments, diagnoses of a kappa light chain homologous recombination.

Generation of mice carrying the kappa chain inactivated

The 4 targeted ES clones described in the previous section were injected into C57Bl / 6J blastocysts as described (Bradley, A. (1987) in Teratocarcinomas and Embryonic Stem Cells: A Practical Approach . EJ Robertson, ed. (Oxford: IRL Press), page 113-151) and transferred to the uterus of pseudopregnated females to generate chimeric mice that represent a mixture of cells derived from the contribution of ES cells and the host blastocyst. Chimeric animals are visually identified by the presence of agouti fur coloration, derived from the ES cell line, on the black background of C57Bl / 6J. ES AB1 cells are an XY cell line, in this way, the male chimeras cross with C57BL / BJ females and the presence of the dominant agouti coat color is controlled in the offspring. The agouti offspring are indicative of the germline transmission of the ES genome. Heterozygosis of the agouti offspring for inactivation of the kappa chain is verified by Southern blot analysis of DNA from tail biopsies using the diagnostic probe used to identify target ES clones. Then, pairings between brothers and sisters of heterozygotes are performed to generate homozygous mice for the mutation of the kappa chain.

Example 12 Inactivation of the Mouse Heavy Chain Gene by Recombination Homologous

This example describes locus inactivation. heavy chain of endogenous murine immunoglobulin by homologous recombination in embryonic stem cells (ES). The strategy is to delete the endogenous heavy chain J segments by homologous recombination with a vector containing sequences of heavy chain in which the J_ {H} region has been deleted and has been replaced by the selectable marker gene neo.

Construction of a heavy chain direction vector

Mouse heavy chain sequences containing the JH region (Figure 26a) were isolated from a phage genomic library derived from the ES D3 cell line (Gossler, et al ., (1986) Proc. Natl. Acad Sci. USA ; 83, 9065-9069) using a JH 4 specific oligonucleotide probe:

5 '- ACT ATG CTA TGG ACT ACT GGG GTC AAG GAA CCT CAG TCA CCG -3 '

A 3.5 SacI / StuI genomic fragment was isolated kb, which covered the JH region, from a positive phage clone and it was subcloned in puc18 digested with SacI / SmaI. Plasmid The result was called puc18 JH. The resistance gene to neomycin (neo), used for drug selection of ES cells transfected, was obtained from plasmid pGEM7 (KJ1). The place HindIII in pGEM7 (KJ1) became a SalI site by the addition of a synthetic adapter, and the neo expression cassette is cleaved by digestion with XbaI / SalI. Then the ends of neo fragment became blunt by treatment with the form Klenow of DNA poly and the neo fragment was subcloned at the site NaeI of puc18 J_ {H}, generating the plasmid puc18 J_ {H} -neo (Figure 26b).

An additional vector construction was performed of address in a derivative of plasmid pGP1b. pGP1b was digested with the restriction enzyme NotI and ligated with the following oligonucleotide as adapter:

5 '- GGC CGC TCG ACG ATA GCC TCG AGG CTA TAA ATC TAG AAG AAT TCC AGC AAA GCT TTG
GC - 3 '

The resulting plasmid, called pGMT, was used to prepare the heavy chain steering construction of mouse immunoglobulin.

The thymidine kinase (TK) gene of herpes simplex virus (HSV) was included in the construction to allow enrichment in ES clones carrying homologous recombinants, as described in Mansour et al . ((1988) Nature 336 , 348-352). The HSV TK gene was obtained from plasmid pGEM7 (TK) by digestion with EcoRI and HindIII. The TK DNA fragment was subcloned between the EcoRI and Hind III sites of pGMT, creating plasmid pGMT-TK (Figure 26c).

To provide a wide region of homology with the target sequence, a genomic XbaI / XhoI fragment of 5.9 kb, located in the 5 'position of the JH region, from a Genomic positive phage clone by limited DNA digestion with XhoI and partial digestion with XbaI. As highlighted in the figures 26a and 26b, this XbaI site is not present in genomic DNA, but which rather derives from the phage sequences that flank directly the cloned genomic heavy chain insert in the clone of positive phage. The fragment was subcloned into pGMT-TK digested with XbaI / XhoI, to generate the plasmid pGMT-TK-J_ {H} 5 '(figure 26d).

The final stage of construction involved the excision of the 3 kb EcoRI fragment from puc18 J_ {H} -neo containing the neo gene and sequences flanking genomics. This fragment became blunt by the Klenow polymerase and was subcloned into the XhoI site of pGMT-TK-J_ {H} 5 'turned into blunt similarly. The resulting construction, J_ {H} K01 (figure 26e), contains 6.9 kb of genomic sequences flanking the locus J_ {H}, with a deletion of 2.3 kb occupying the region J_ {H} in which the neo gene has been inserted. Figure 25f shows the structure of an endogenous heavy chain allele after the homologous recombination with the address construct.

       \ newpage
    

Example 13 Generation and analysis of target ES cells

ES AB-1 cells (McMahon, AP and Bradley, A. (1990) Cell 62 , 1073-1085) were cultured in layers of mitotically inactive SNL76 / 7 nurse cells essentially as described (Robertson, EJ (1987) Teratocarcinomas and Embryonic Stem Cells: A Practical Approach. EJ Robertson, ed. (Oxford: IRL Press), p. 71-112).

The JH K01 heavy chain inactivation vector was digested with NotI and introduced by electroporation into AB-1 cells by the methods described (Hasty, PR, et al . (1991) Nature 350 , 243-246). The electroporated cells were plated at 100 mm at a density of 2-5 x 10 6 cells / plate. After 24 hours, G418 (200 mg / ml active component) and FIAU (0.5 mM) were added to the medium, and drug resistant clones were allowed to develop for 8-10 days. Clones were chosen, trypsinized, divided into two portions and further expanded. Then, half of the cells obtained from each clone were frozen and homologous recombination between the vector and target sequences in the other half of the cells was analyzed.

DNA analysis is performed by Southern blot hybridization. DNA is isolated from the clones as described (Laird, PW et al ., (1991) Nucl. Acids Res ., 19 ), digested with HindIII and probed with the 500 bp EcoRI / StuI fragment called probe of diagnosis in Figure 26f. This probe detects a 2.3 kb HindIII fragment in the wild-type locus, while a 5.3 kb band is diagnostic of a target locus that has homologously recombined with the direction vector (see Figures 26a and f ). Additional digestions are made with the SpeI, StuI and BamHI enzymes to verify the directed breakage of the heavy chain allele.

Example 14 Heavy Chain Minilocus Transgen A. Construction of plasmid vectors for cloning of large DNA sequences 1. pGP1a

Plasmid pBR322 was digested with EcoRI and with StyI and ligated with the following oligonucleotides:

22

The resulting plasmid, pGP1a, is designed to clone very large DNA constructs that can be cleaved by the rare cut restriction enzyme NotI. It contains a downstream NotI restriction site (in relation to the ampicillin resistance gene, AmpR) of a strong transcription termination signal derived from the trpA gene (Christie, GE et al . (1981) Proc. Natl. Acad. Sci. USA , 78 , 4180). This termination signal reduces the potential toxicity of coding sequences inserted into the NotI site by eliminating transcription transcription from the AmpR gene. In addition, this plasmid is of low copy number in relation to the pUC plasmids, since it retains the control region of the copy number of pBR322. The low copy number further reduces the potential toxicity of inserted sequences and reduces the selection against large inserts due to DNA replication.

2. pGP1b

pGP1a was digested with NotI and bound with the following oligonucleotides:

2. 3

       \ newpage
    

The resulting plasmid, pGP1b, contains a short Short polylinker region flanked by NotI sites. This facilitates the construction of large inserts that can be split by digestion with NotI.

3. pGPe

The following oligonucleotides:

24

were used to amplify the immunoglobulin heavy chain 3 'enhancer (S. Petterson, et al . (1990) Nature, 344, 165-168) from rat liver DNA by the polymerase chain reaction technique .

The amplified product was digested with BamHI and with SphI and was cloned in pNN03 digested with BamHI / SphI (pNN03 is a plasmid derived from pUC containing a polylinker with the following restriction sites, listed in order: NotI, BamHI, NcoI, ClaI, EcoRV, XbaI, SacI, XhoI, SphI, PstI, BglII, EcoRI, SmaI, KpnI, HindIII and NotI). The resulting plasmid, pRE3, was digested with BamHI and with HindIII, and the insert containing the 3 'enhancer of the rat Ig heavy chain was cloned into pGP1b digested with BamHI / HindIII. The resulting plasmid, pGPe (Figure 27 and Table 1), it contains several unique restriction sites in which they can clone sequences and subsequently cleave together with the 3 'enhancer by digestion with NotI.

       \ vskip1.000000 \ baselineskip
    

       \ vskip1.000000 \ baselineskip
    

       \ vskip1.000000 \ baselineskip
    

(Table goes to page next)

TABLE 1 PGPe vector sequence

25

B. Construction of the minilocus transgene that expresses IgM, pIGM1 1. Isolation of constant region clones J- \ mu and construction of pJM1

A genomic DNA library of human placenta cloned into the phage vector λEMEM3 /
SP6 / T7 (Clonetech Laboratories, Inc., Palo Alto, CA) with the human heavy chain J region specific oligonucleotide:

26

and the phage clone was isolated λ1.3. A HindIII / KpnI fragment was isolated from this clone of 6 kb, which contained the six J segments as well as the DHQ52 D segment and the heavy chain J- \ intronic enhancer. The same library was analyzed with the specific oligonucleotide of \ mu human:

27

and the phage clone was isolated λ2.1. A fragment was isolated from this clone 10.5 kb HindIII / XhoI containing the change region µ and all exons of the constant region µ. These two fragments were ligated together with pNN03 digested with KpnI / XhoI to get the plasmid pJM1.

2. pJM2

A 4 kb XhoI fragment was isolated from phage clone λ2.1 containing sequences immediately downstream of the sequences in pJM1, including the so-called \ Sigma \ element involved in the deletion of µ in certain B cells that express IgD (H. Yasui et al . (1989) Eur. J. Immunol . 19 , 1399). This fragment was treated with the Klenow fragment of DNA polymerase I and ligated with pJM1 cut with XhoI and treated with Klenow. The resulting plasmid, pJM2 (Figure 28), had lost the internal XhoI site but retained the 3 'NhoI site due to the incomplete reaction of the Klenow enzyme. pJM2 contains the complete human J region, the heavy chain intronic enhancer J- µ, the change region µ and all the exons of the constant region µ, as well as the two direct 0.4 kb repeats, ma sigma \ mu and \ Sigma \ mu, involved in the deletion of \ mu.

3. Isolation of clones from region D and construction of pDH1

The following specific oligonucleotide of the human D region:

28

to select clones from region D of the genomic library of human placenta. Phage clones λ4.1 and λ4.3 were isolated. A 5.5 kb XhoI fragment was isolated, which included the elements D, D 11, D N1 and D M2 (Y. Ichihara et al . (1988) EMBO J. , 7 , 4141) from phage clone λ4.1. An adjacent fragment upstream of 5.2 kb was isolated, which included the elements D, D_ {LR1}, D_ {XP1}, D_ {XP-1} and D_ {A1}, from the phage clone λ4. 3. Each of these XhoI region D fragments was cloned into the Sa1I site of the plasmid vector pSP72 (Promega, Madison, WI) to thereby destroy the XhoI site that linked the two sequences. The upstream fragment was then cleaved with XhoI and SmaI, and the downstream fragment with EcoRV and XhoI. The resulting isolated fragments were ligated with pSP72 digested with SalI to give plasmid pDH1. pDH1 contains a 10.6 kb insert that includes at least 7 D segments and can be cleaved with XhoI (5 ') and EcoRV (3').

4. pCOR1

Plasmid pJM2 was digested with Asp718 (a KpnI isosquizomer) and the projection was filled with the fragment Klenow of DNA polymerase I. Next, the resulting DNA was digested with ClaI and the insert was isolated. This insert was linked with the insert XhoI / EcoRV of pDH1 and with pGPe digested with XhoI / ClaI to generate pCOR1 (figure 29).

5. pVH251

A 10.3 kb HindIII genomic fragment containing the two human heavy chain variable region segments VH 251 and VH 105 (CG Humphries et al . (1988) Nature 331, 446) was subcloned into pSP72 to give plasmid pVH251.

6. pIGM1

Plasmid pCOR1 was partially digested with XhoI and the XhoI / SalI insert isolated from pVH251 was cloned into the site XhoI upstream to generate plasmid pIGM1. pIGM1 contains 2 functional human variable region segments, at least 8 human D segments, the 6 human JH segments, the enhancer of human J- \ mu, the human \ sigma \ mu element, the region of human change, all the exons that encode \ mu human and the element \ Sigma \ mu human, together with the 3 'rat heavy chain enhancer, so that all these sequence elements can be isolated in a single fragment to from vector sequences by digestion with NotI and microinject into pronuclei of mouse embryos to generate transgenic animals

       \ newpage
    

C. Construction of the minilocus transgene that expresses IgM and IgG, pHC1 1. Isolation of constant region clones γ

The following oligonucleotide, specific for g region genes of human Ig:

29

was used to analyze the library human genomics Phage clones 129.4 and λ29.5. A 4 kb HindIII fragment of the phage clone λ29.4, which contained a γ change region, was used to probe a genomic DNA library of human placenta cloned into the lambda phage vector FIX? II (Stratagene, La Jolla, CA). Phage clone? LambSg1.13 was isolated. For determine the subclass of the different γ clones, it performed dideoxy sequencing reactions using subclones of each of the three phage clones as a mold and the next oligonucleotide as primer:

oligo-67 5'-tga gcc cag aca ctg gac -3 '

It was determined that the two phage clones λ29.5 and λS γ1.13 were subclass γ1.

2. p γ

A 7.8 kb HindIII fragment of the phage clone λ29.5, which contained the coding region γ1, was cloned in pUC18. The resulting plasmid, pLT1, was digested with XhoI, He treated himself with Klenow and became religious to destroy the internal XhoI site. The resulting clone, pLT1xk, was digested with HindIII, the inserted and cloned into pSP72 to generate plasmid clone pLT1xks. Digestion of pLT1xks at an XhoI site of the polylinker and in a BamHI site derived from human sequence generates a 7.6 fragment kb containing the exons that encode the constant region γ1. This 7.6 kb XhoI / BamHI fragment was cloned in conjunction with an adjacent 4.5 kb BamHI fragment chain below phage clone λ29.5 in pGPe digested with XhoI / BamHI to generate the plasmid clone p γ. p γ contains all exons encoding the constant region γ1, together with 5 kb of downstream sequences, linked to the 3 'enhancer of heavy rat chain

3. p γ2

A 5.3 kb HindIII fragment containing the γ1 change region and the first exon of the unproductive transcript prior to the change (P. Sideras et al . (1989) International Immunol . 1 , 631) was isolated from the clone of phage λSγ1.13 and was cloned into pSP72 with the XhoI site of the polylinker adjacent to the 5 'end of the insert, to generate plasmid clone pSγ1s. The XhoI / SalI insert of pSγ1s was cloned into pγamma1 digested with XhoI to generate the plasmid clone pγ2 (Figure 31). p γ2 contains all exons encoding the γ1 constant region, and exons of the upstream and unproductive transcript region, together with 5 kb of downstream sequences, attached to the 3 'heavy chain rat enhancer. This clone contains a unique XhoI site at the 5 'end of the insert. The complete insert, together with the XhoI site and the 3 'rat enhancer, can be cleaved from the sequences of the vector by digestion with NotI.

4. pHC1

Plasmid pIGM1 was digested with XhoI and the 43 kb insert was isolated and cloned into pge2 digested with XhoI to generate plasmid pHC1. pHC1 contains 2 region segments Functional human variable, at least 8 human D segments, all 6 human segments J_ {H}, the enhancer J- \ mu human, the element \ sigma \ mu human, the region of change \ mu human, all exons coding for the human region, the human \ Sigma \ mu element and the constant region γ1 human, including the region of associated change and exons associated with unproductive transcription, together with the 3 'heavy chain rat enhancer, so that all these sequence elements can be isolated in a single fragment from of the vector sequences by digestion with NotI and microinject into mouse embryo pronuclei to generate transgenic animals

D. Construction of the minilocus transgene that expresses IgM and IgG, pHC2 1. Isolation of the heavy chain V region gene human VH49.8

The lambda library of genomic placenta DNA Human (FIX? II, Stratagene, La Jolla, CA) was analyzed with the following specific oligonucleotide of the human VH1 family:

30

Phage clone λ4949 was isolated and a 6.1 kb XbaI fragment containing the VH49.8 variable segment was subcloned into pNNO3 (so that the ClaI site of the polylinker is chain down VH49.8 and the XhoI site of the polylinker is chain above) to generate plasmid pVH49.8. A region of 800 bp of this insert and it was discovered that VH49.8 had a phase of open reading and ayuste and recombination signals intact, indicating in this way that the gene is functional (table 2).

TABLE 2 Sequence of the V_ {H} 49.8 gene of the V_ {H} I family human

31

2. pV2

A 4 kb XbaI genomic fragment containing the VH 4-21 gene of the human VH IV family (I. Sanz et al . (1989) EMBO J. , 8 , 3741), subcloned into the plasmid pUC12, was cleaved with SmaI and HindIII, and treated with the Klenow fragment of polymerase I. Next, the blunt end fragment was cloned into pVH49.8 digested with ClaI and treated with Klenow. The resulting plasmid, pV2, contains the VH49.8 human heavy chain gene linked upstream of VH4-21 in the same orientation, with a single SalI site at the 3 'end of the insert and a single XhoI site at the 5' end .

3. pS γ1-5 '

A 0.7 kb XbaI / HindIII fragment (representing sequences immediately upstream, and adjacent, to the fragment containing the 5.3 kb shift region in plasmid pγ2) together with the XbaI 3 fragment , 1 kb upstream chain were isolated from the phage clone λSg1.13 and cloned into the pUC18 vector digested with HindIII / XbaI. The resulting plasmid, pSγ1-5 ', contains a 3.8 kb insert representing sequences upstream of the unproductive transcription initiation site found in B cells before the change to the γ1 isotype (P. Sideras et al . ( 1989) International Immunol., 1, 631). As the transcript is involved in the initiation of the isotype change, and cis-acting sequences are often important for transcription regulation, these sequences are included in the transgenic constructs to promote the correct expression of the unproductive transcript and recombination. of associated change.

4. pVGE1

The pS γ1-5 'insert is cleaved with SmaI and with HindIII, treated with the enzyme Klenow and ligated with the following oligonucleotide linker:

5 '- ccg gtc gac cgg - 3'

The ligation product was digested with SalI and was ligated to pV2 digested with SalI. The resulting plasmid, pVP, Contains 3.8 kb of 5 'flanking sequences of change γ1 linked down the two human variable gene segments VH49.8 and VH4-21 (see table 2). PVP insert it is isolated by partial digestion with SalI and complete digestion with XhoI, followed by purification of the 15 kb fragment on a gel agarose Then, the insert is cloned into the XhoI site of p γ2 to generate the plasmid clone pVGE1 (Figure 32). pVGE1 It contains two variable human heavy chain gene segments upstream of the human γ1 constant gene and the region of associated change. A single SalI site can be used between regions variable and constant for cloning in gene segments D, J and \ mu. The 3 'rat heavy chain enhancer is attached to the 3 'end of the γ1 gene and the entire insert is flanked by NotI sites.

5. pHC2

Plasmid clone pVGE1 is digested with SalI and the XhoI insert of pIGM1 is cloned into it. The resulting clone, pHC2, contains 4 functional human variable region segments, at least 8 human D segments, the 6 human J H segments, the human J-m enhancer, the element \ sigma \ mu human, the region of change \ mu human, all exons that encode human, the human element and the region human γ1 constant, including the associated region of change and exons associated with unproductive transcription, in conjunction with 4 kb flanking sequences upstream of the site of initiation of unproductive transcription. These human sequences are join the 3 'heavy rat rat chain enhancer, so that all the sequence elements can be isolated in a single fragment to from vector sequences by digestion with NotI, and microinject into mouse embryo pronuclei to generate transgenic animals A single XhoI site can be used in the 5 'end of the insert for cloning into gene segments additional human variables to further expand the Recombinant diversity of this heavy chain minilocus.

E. Transgenic mice

NotI inserts of plasmids pIGM1 and pHC1 were isolated from vector sequences by agarose gel electrophoresis. The purified inserts were microinjected into the pronuclei of mouse embryos (C57BL / 6 x CBA) F2 fertilized and the surviving embryos were transferred to pseudopregnated females as described in Hogan et al . (B. Hogan, F. Costantini and E. Lacy, Methods of Manipulating the Mouse Embryo, 1986, Cold Spring Harbor Laboratory, New York). In mice that developed from the injected embryos, the presence of transgenic sequences was analyzed by Southern blot analysis of the tail DNA. The number of copies of the transgene was estimated by the intensity of the band in relation to control standards containing known amounts of cloned DNA. Between 3 and 8 weeks of age, serum from these animals was isolated and the presence of human IgM and IgG1 encoded by the transgene by ELISA was tested as described by Harlow and Lane (E. Harlow and D. Lane. Antibodies: A Laboratory Manual, 1988, Cold Spring Harbor Laboratory, New York). Microtiter plates wells were coated with mouse monoclonal antibodies specific for human IgM (clone AF6, = 0285, AMAC, Inc. Westbrook, ME) and for human IgG1 (clone JL512, = 0280, AMAC, Inc. Westbrook, ME) . Serum samples were serially diluted in the wells and the presence of specific immunoglobulins with goat anti-human Ig (polyvalent) conjugated with alkaline phosphatase and isolated by affinity that had been previously adsorbed to minimize cross reactivity with immunoglobulins of mouse. Figure 33 shows the results of an ELISA test for the presence of human IgM and IgG1 in the serum of two animals developed from embryos injected with the transgenic insert of plasmid pHC1. One of the animals (# 18) was negative for transgene by Southern blot analysis, and showed no detectable levels of human IgM or IgG1. The second animal (# 38) contained approximately 5 copies of the transgene, as assessed by Southern blotting, and showed detectable levels of both human IgM and IgG1. The results of the ELISA tests of 11 animals developed from embryos in which the transgene had been injected are summarized in the following table (table 3).

TABLE 3 Detection of human IgM and IgG1 in serum transgenic animals by ELISA test

Nº copies approximate No. of animal injected transgen of the transgender (per cell) Human IgM IgG1 human 6 pIGM1 one ++ - 7 pIGM1 0 - - 9 pIGM1 0 - - 10 pIGM1 0 - - 12 pIGM1 0 - - fifteen pIGM1 10 ++ - 18 pHC1 0 - - 19 pHC1 one - - twenty-one pHC1 < one - - 26 pHC1 2 ++ + 38 pHC1 5 ++ +

Table 3 shows a correlation between the presence of integrated transgenic DNA and the presence of immunoglobulins encoded by the transgene in the serum. Two of the animals that, as discovered, contained the pHC1 transgene not expressed detectable levels of human immunoglobulins. Both of them they were animals with a low number of copies and may not contained complete copies of transgenes, or animals could be genetic mosaics (indicated by the <1 estimated copy for animal # 21), and the cells that contain the transgene They may not have populated the hematopoietic lineage. So alternatively, transgenes may have been integrated into Genomic locations that do not lead to its expression. The detection of human IgM in the serum of transgenic pIGM1, and of IgM and human IgG1 in the transgenic pHC1, indicates that the sequences transgenics work correctly to direct the VDJ junction, the transcription and isotype change.

Example 15 Reorganized Heavy Chain Transgenes A. Isolation of Human Heavy Chain VDJ Segments Reorganized

Two leukocyte genomic DNA libraries human cloned into phage vector 1EMBL3 / SP6 / T7 (Clonetech Laboratories, Inc., Palo Alto, CA) are analyzed with a fragment PacI / HindIII of λ1.3 of 1 kb containing the enhancer J- \ human heavy chain intronic. The Positive clones are assayed by hybridization with a mixture of the following VH-specific oligonucleotides:

32

The clones that hybridize with the two probes V and J- µ are isolated and the DNA sequence is determined of the reorganized VDJ segment.

B. Construction of human heavy chain transgenes reorganized

Fragments containing segments are subcloned Functional VJ (open reading phase and help signals) in the plasmid vector pSP72 so that the XhoI site derived from the Plasmid is adjacent to the 5 'end of the insert sequence. A subclone containing a functional VDJ segment is digested with XhoI and PacI (PacI, an unusual cutting enzyme, which recognizes a site next to the intronic enhancer J-m) and the insert is cloned in pHC2 digested with XhoI / PacI to generate a transgenic construction with a functional VDJ segment, the J- \ intronic enhancer, the element of change µ, the exons encoding the constant region µ and the constant region γ1, including the sequences associated with the unproductive transcript, the change sequence γ1 and the coding exons. This transgenic construction is cleaved with NotI and is microinjected into the mouse embryo pronuclei to Generate transgenic animals as described above.

Example 16 Light Chain Transgenes A. Construction of plasmid vectors 1. Plasmid vector pGP1c

Plasmid vector pGP1a is digested with NotI and binds with the following oligonucleotides in:

33

The resulting plasmid, pGP1c, contains a polylinker with restriction sites XmaI, XhoI, SalI, HindIII and BamHI flanked by NotI sites.

2. Plasmid vector pGP1d

Plasmid vector pGP1a is digested with NotI and binds with the following oligonucleotides in:

3. 4

The resulting plasmid, pGP1d, contains a polylinker with restriction sites SalI, HindIII, ClaI, BamHI and XhoI flanked by NotI sites.

B. Isolation of clones J \ kappa and C \ kappa

A genomic DNA library of human placenta cloned into phage vector \ lambdaEMBL3 / SP6 / T7 (Clonetech Laboratories, Inc. Palo Alto, CA) was analyzed with the oligonucleotide Region specific J of human kappa light chain:

35

and phage clones 136.2 and 136.5 were isolated. A 7.4 kb XhoI fragment that included the J? Segment is isolated from 136.2 and subcloned into plasmid pNN03 to generate plasmid clone p36.2. A nearby 13 kb XhoI fragment which includes segments JK from 2 to 5 together with the segment C? gene was isolated from phage clone 136.5 and was subcloned into plasmid pNN03 to generate plasmid clone p36.5. These two clones together include the region that starts 7.2 kb string up J \ kappa1 and finish 9 kb string down C \ kappa.

C. Construction of light chain transgenes reorganized 1. pCK1, a C \ kappa vector to express segments reorganized variables

The 13 kb XhoI insert of the plasmid clone p36.5 containing the C? gene, in conjunction with 9 kb of downstream sequences, it is cloned into the SalI site of the vector plasmid pGP1c with the 5 'end of the insert adjacent to the site XhoI of the plasmid. The resulting clone pCK1 can accept fragments cloned containing VJ \ kappa segments reordered on site unique XhoI 5 '. Then, the transgene can be cleaved with NotI and purified from the vector sequences by electrophoresis in gel The resulting transgenic construction will contain the human J-C \ kappa intronic enhancer and can contain the human 3 '\ kappa enhancer.

2. pCK2, a C \ kappa vector with chain enhancers heavy to express reorganized variable segments

A 0.9 kb XbaI fragment of mouse genomic DNA containing the mouse heavy chain J- µ intronic enhancer (J. Banerji et al . (1983) Cell 33, 729-740) was subcloned into pUC18 to generate plasmid pJH22.1. This plasmid was linearized with SphI and the ends were filled with the enzyme klenow. Then, the klenow-treated DNA was digested with HindIII and ligated to a MluI (klenow) / HindIII fragment of phage clone λ1.3 (above example), which contained the human heavy chain J-potencia intronic enhancer ( A. Hayday et al ., (1984) Nature 307, 334-340). The resulting plasmid, pMHE1, consisting of the mouse and human heavy chain intronic enhancer J- µ was ligated together with pUC18 so that they were cleaved into a single BamHI / HindIII fragment. This 2.3 kb fragment was isolated and cloned into pGP1c to generate pMHE2. pMHE2 was digested with SalI and in it the XhoI insert of 13 kb p36.5 was cloned. The resulting plasmid, pCK2, is identical to pCK1, except that the J-µ mouse and human heavy chain intronic enhancers are fused to the 3 'end of the transgene insert. To modulate the expression of the final transgene, analogous constructions can be generated with different enhancers, that is, the 3 'kappa or mouse or rat heavy chain enhancer (K. Meyer and MS Neuberger, (1989) EMBO J. , 8 , 1959-1964; S. Petterson, et al . (1990) Nature , 344 , 165-168).

2. Isolation of variable light chain segments reorganized kappa

Two human leukocyte genomic DNA libraries cloned in the phage vector λEMEM3 / were analyzed
SP6 / T7 (Clonetech Laboratories, Inc., Palo Alto, CA), containing the human Kappa light chain J region the XhoI / SmaI fragment of 3.5 kb p36.5. Hybridization of the positive clones was tested with the following V? Specific oligonucleotide:

36

Clones that hybridized with both were isolated probe V as with probe J and the DNA sequence was determined of the reorganized VJ \ kappa segment.

3. Generation of transgenic mice containing reorganized human light chain constructions

Fragments containing segments are subcloned Functional VJ (open reading phase and assist signals) in the Unique XhoI sites of vectors pCK1 and pCK2 to generate reorganized kappa light chain transgenes. The constructions transgenic are isolated from vector sequences by digestion with NotI. The agarose gel purified insert is microinjects into mouse embryo pronuclei to generate transgenic animals Animals expressing human kappa chain intersect with transgenic animals that contain minilocus of heavy chain (example 14) to generate mice that express completely human antibodies.

As not all VJ \ kappa combinations can be able to form chain complexes heavy-light stable with a broad spectrum of different heavy chain VDJ combinations, several are generated different light chain transgenic constructions, using each a different reorganized VJ \ clone, and mice transgenic results of these constructs intersect with mice that express the heavy chain minilocus transgene. They are isolated peripheral blood, spleen and lymph node lymphocytes double transgenic animals (with both chain constructions light as for heavy chain), stained with antibodies specific fluorescent for heavy and light chains of human and mouse immunoglobulins (Pharmingen, San Diego, CA) and se analyze by flow cytometry using a FACScan analyzer (Becton Dickinson, San Jose, CA). For the generation of antibodies human monoclonal, transgene constructs of reorganized light chain that produce the highest level of complexes of human heavy / light chains on the surface of the greater number of B cells, and do not adversely affect the immune cell compartment (assayed by cytometry analysis flow with antibodies specific for subpopulations of B and T cells).

D. Construction of chain minilocus transgenes lightweight not rearranged 1. pJCK1, a vector containing J \ kappa and C \ kappa to build minilocus transgenes

The XhoI insert of 13 kb p36.5 containing C? Is treated with the enzyme klenow and cloned into the plasmid pGP1d digested with HindIII and treated with klenow. A plasmid clone so that the 5 'end of the insert is adjacent to the ClaI site derived from the vector. The resulting plasmid, p36.5-1d, it is digested with ClaI and treated with klenow. Next, the 7.4 kb XhoI insert of p36.2 containing J \ kap1 It is treated with klenow and is cloned in p36.5-1d treated with ClaI and with klenow. A clone is selected in which the insert p36.2 is in the same orientation as the p36.5 insert. This clone, pJCK1 (Figure 34), contains the complete human J \ kappa region and C κ, together with 7.2 kb of upstream and 9 sequences kb of downstream sequences. The insert also contains the human J-C \ kappa intronic enhancer and can contain a 3 '\ kappa human enhancer. The insert is flanked by a single SalI 3 'site in order to clone additional 3 'flanking sequences such as enhancements of heavy chain or light chain. A unique XhoI site is located at the 5 'end of the insert in order to clone segments therein non-rearranged V? gene. The unique sites SalI and XhoI are flanked in turn by NotI sites that are used to isolate the complete transgenic construction of the sequences of the vector.

2. Isolation of V \ kappa gene segments no reorganized and generation of transgenic animals that express light chain protein of human Ig.

V \ kappa specific oligonucleotide, oligo-65 (described above), is used as a probe from a genomic DNA library of human placenta cloned into the phage vector 1EMBL3 / SP6 / T7 (Clonetech Laboratories, Inc. Palo Alto, CA). The variable gene segments of the resulting clones and clones that appear to be selected are selected functional. Criteria for judging functionality include: open reading phases, acceptor and donor sequences of ayuste intact and recombination sequence intact. Fragments are cloned of DNA containing variable gene segments selected in the unique XhoI site of plasmid pJCK1 to generate constructs of minilocus The resulting clones are digested with NotI, they are isolated the inserts and injected into mouse embryo pronuclei to Generate transgenic animals. The transgenes of these animals they will experience V to J binding in developing B cells. The animals expressing human kappa chain intersect with animals transgenic containing the heavy chain minilocus (example 14) to generate mice that express antibodies completely humans.

Example 17 Variable Synthetic Heavy Chain Region

This example is summarized in Figure 35.

A. Construction of the pVHf Cloning Vector 1. pGP1f

Plasmid pGP1a (previous example) is digested with NotI and the following oligonucleotides bind with it:

37

The resulting plasmid, pGP1f, contains sites SphI, XhoI and HindIII flanked by NotI and SfiI sites.

2. pVHf

The VH 251 gene segment of the human VH-V family (CG Humphries et al . (1988) Nature , 331 , 446) together with approximately 2.4 kb of 5 'and approximately 1 flanking sequences, 4 kb of 3 'flanking sequences, was isolated on a 4.2 kb SphI / HindIII fragment of plasmid clone pVH251 (previous example) and cloned into plasmid vector pSelect ™ -1 (Promega Corp. Madison, WI ). The 5 'flanking sequences, together with the promoter, the first exon and the first intron of VH 251, are amplified by polymerase chain reaction (PCR) from this template using the following oligonucleotides:

38

3 'flanking sequences are amplified by PCR using the following oligonucleotides:

The 5 'amplified sequences are digested with SphI and with XhoI, and the 3 'sequences are digested with HindIII and with XhoI. The resulting fragments are cloned together in the plasmid pGP1f to generate plasmid pVHf. PVHf plasmid It contains the cis acting regulating elements that control the transcription of VH 251, together with the signal sequence which encodes the first exon. pVHf is used as a cassette of expression for heavy chain variable sequences. These sequences are cloned into the plasmid digested with KasI / XhoI as describe later.

B. Isolation of Gene Coding Sequences Variable 1. Amplification of cDNA sequences of VH genes expressed

Poly (A) + RNA is isolated from human peripheral blood lymphocytes (PBL). The first cDNA chain is synthesized with reverse transcriptase, using oligo- (dT) as the primer. The first cDNA chain is isolated and a tail with oligo (dG) is introduced using the terminal transferase. Then, the 5 'sequences of the IgM transcripts are specifically amplified by a modification of the method of Frohman et al ., (1988, Proc. Natl. Acad. Sci. USA , 85 , 8998). Oligo- (dC) 13 and the following oligonucleotide:

oligo-69 5 '-gga att ctc aca gga gac gag - 3 '

they are used as primers 5 'and 3', respectively, in a polymerase chain reaction with the first PBL cDNA chain with dG tail. He oligo-69 is complementary to sequences that encode amino acids 11-17 of the domain IgM constant. Therefore, these primers will amplify approximately 0.6 kb DNA fragments that include sequences of VH genes expressed.

2. Retroconversion of cDNA sequences in the form of germ line

The following oligonucleotide:

39

hybridizes with 5 'IgM sequences PCR amplified and denatured. The oligo- "c" includes a non-degenerated 21 nucleotide sequence that includes a site KasI, followed by a degenerate sequence of 30 nucleotides that is homologous to the 5 'end of the second exon of many segments V H humans (Genbank; Los Alamos, NM). The primer is prolonged with the DNA polymerase and the product is isolated from the primer not used by fractionation by size. Then the product is denatured and hybridize with the following oligonucleotide:

40

The oligo- "d" includes a sequence not degenerate of 30 nucleotides that includes an XhoI site and part of the V recombination sequence with DJ, followed by a sequence degenerate of 21 nucleotides that is complementary to the sequence which encodes the last seven amino acids of the structural region three of many human variable gene segments. Later prolongs the oligonucleotide hybridized with DNA polymerase and the product is isolated from unused primer by fractionation by size. Individual cycles of DNA synthesis are performed followed of the elimination of the primers to ensure the integrity of the sequence of individual variable gene fragments. The product of the oligo- "d" primer extension is amplified by PCR using the following two oligonucleotides as primers:

41

The resulting 0.36 kb PCR product is Purifies by gel electrophoresis and is digested with the enzymes of KasI and XhoI restriction. Then the digestion products will cloned in pVHf digested with KasI / XhoI to generate a library of variable gene sequences expressed in line configuration germinal. The link on the KasI site of pVHF reproduces the site ayuste acceptor of the 5 'end of the second exon, while the linkage at the XhoI site reproduces the recombination signal of the 3 'end of the variable gene segment. Versions are used degenerate oligonucleotide alternatives "c" and "d" to amplify different populations of variable genes and to generate libraries in germline configuration that they represent the different populations (Genbank; Los Alamos, NN).

C. Construction of Synthetic Locus

The complete gene library is developed Synthetic V_ {H} in germinal configuration together and isolate plasmid DNA. The average number pVHf plasmid of copies, which includes a powerful transcription terminator between the ampicillin resistance gene and the cloning site, is designs to minimize the expansion of particular clones within library. Plasmid DNA is digested with SfiI, treated with calf intestinal phosphatase to remove 5 'phosphate groups, and then digested with NotI. Calf intestinal phosphatase is removed before digestion with NotI, so that only dephosphorylate the SfiI ends. Then the digested DNA is isolated from the vector sequences by agarose gel electrophoresis and it binds with the following oligonucleotides:

42

The oligo- "h" is phosphorylated while The oligo- "g" is not phosphorylated on the left. The reaction of ligation is performed with a large molar excess of oligonucleotides so that all the NotI ends of the gene fragment V are will bind with oligonucleotides and not with other region fragments V. Because the SfiI ends are not compatible with each other, the V segments will be concatenated in the same orientation, so that each segment V will be separated from the next segment V by a single oligonucleotide spacer unit.

Large concatomers are separated by size in electrophoresis and are isolated in agarose gels. Then the size fractionated concatomers are injected together directly in mouse embryo pronuclei in conjunction with the DNA fragments containing D-J-C (such as inserts pHC1 or pHC2) to generate transgenic animals with repertoires large primaries Alternatively, concatomers are cloned in a plasmid vector such as pGPf.

Example 18 Generation of Specific Antibodies of Subpopulations of Lymphoid Cell Receptors

Inoculation of mice with immunoglobulins (xenogeneic (ie, human) B cell receptors or with T cell receptors predominantly leads to the generation of mouse antibodies directed against particular epitopes (dominant epitopes) shared by all or most of the immunoglobulins or T-cell receptors of a species determined, but that differ between species. Therefore it is difficult to isolate antibodies that distinguish between subpopulations particular of B or T cell receptors (eg idiotypes or families of variable region). However, the transgenic mouse expressing human immunoglobulins (described in the examples above) will be immunologically tolerant to epitopes of shared B cells and therefore will be useful for generation of antibodies that distinguish between subpopulations of human immunoglobulins This concept extends to the generation of transgenic mice expressing sequences encoding the human T cell receptor and the crossing of these mice with the Transgenic mouse for human immunoglobulin. In these mice inoculate isolates that contain T cell receptor proteins human and monoclonal antibodies are generated that recognize T cell receptor subpopulations

Certain studies have shown that there is limited variability of T-cell antigen receptors implicated in certain autoimmune diseases (TF Davies et al . (1991) New England J. Med ., 325 , 238). Due to this limited variability, it is possible to generate human monoclonal antibodies that specifically recognize the subpopulation of human T cells that is self reactive.

A. Generation of specific antibodies from B cell subpopulations

In transgenic mice expressing human immunoglobulin, isolated immunoglobulins are inoculated from a healthy donor or a patient with a B-cell malignancy expressing a high level of a single type of immunoglobulin (Miller et al ., (1982) New Eng. J. Med. 306, 517-522). Monoclonal antibody secretory hybridomas are generated as described by Harlow and Lane (E. Harlow and D. Lane. Antibodies: A Laboratory Manual, 1988. Cold Spring Harbor Laboratory, New York). Individual hybridomas that secrete human antibodies that specifically recognize subpopulations of B cells are selected.

B. Transgenic mice expressing sequences of the human T cell receptor

DNA fragments containing the α and β genes of the human T cell receptor (TCR) intact and completely rearranged into mouse embryo pronuclei are generated to generate transgenic animals. The expression in the transgenic animals of both transgenes on the surface of their T cells is tested by FACS analysis. Animals expressing only low levels of α and β TCR chains in a fraction of their T cells are selected. Only low levels of expression are required to obtain immunological tolerance, and a high level of expression could destabilize the immune system of the animal and interfere with the ability to generate an immune response required for the generation of monoclonal antibodies. Alternatively, because a correct tissue or cell-specific expression is not required to obtain immunological tolerance, cDNA clones of α and β TCR chains are inserted into transgenic expression cassettes (T. Choi et al. ., (1991) Mol. Cell. Biol ., 11 ; 3070-3074) under the control of transcription signals other than those of TCR. The transgenic constructs of cDNA of α and β chains of TCR are injected together into mouse embryo pronuclei to generate transgenic animals. Ectopic expression of TCR chains will not result in cell surface expression, since TCR is a multi-chain complex (H. Clevers et al ., 1988 Ann. Rev. Immunol ., 6, 629-662); however, expression on the cell surface is not required for antigen presentation (Townsend et al ., (1986) Nature , 324 , 575-577) or for tolerance induction.

Transgenic mice cross for chains α and β of the T cell receptor with mice transgenics expressing human immunoglobulin to generate mice useful for generating human monoclonal antibodies that recognize specific subpopulations of human T cells. These mice are inoculated with proteins derived from isolated T cells at from a healthy donor or a patient with a malignancy of T cells expressing a single type of TCR. Hybridomas are generated that secrete monoclonal antibodies and hybridomas are selected individuals that secrete human antibodies that recognize specifically subpopulations of B cells.

Example 19 Genomic Transgen of Human Ig Heavy Chain

This example describes the cloning of a transgene genomic human immunoglobulin heavy chain which is then introduced into the murine germ line by means of microinjection in zygotes or by integration into ES cells.

Nuclei are isolated from newly obtained human placenta tissue as described by Marzluff, WF, et al . (1985), Transcription and Translation: A Practical Approach , BD Hammes and SJ Higgins, eds, p. 89-129, IRL Press, Oxford). Isolated nuclei (or human sperm washed with PBS) are included in 0.5% low melting point agarose blocks and lysed with 1 mg / ml K proteinase in 500 mM EDTA, 1% SDS for nuclei , or with proteinase K at 1 mg / ml in 500 mM EDTA, 1% SDS, 10 mM DTT for sperm at 50 ° C for 18 hours. Proteinase K is inactivated by incubating the blocks in PMSF at 40 µg / ml in TE for 30 minutes at 50 ° C, and then extensively washing with TE. Then, the DNA in the agarose is digested with the restriction enzyme NotI as described by M. Finney in Current Protocols in Molecular Biology (F. Ausubel et al ., Eds. John Wiley & Sons, supl. 4, 1988, for example, Section 2.5.1).

The DNA digested with NotI is then fractionated by pulsed field gel electrophoresis as described by Anand, R. et al . (1989), Nucl. Acids Res ., 17 , 3425-3433. Enriched fractions in the NotI fragment are assayed by Southern hybridization to detect one or more of the sequences encoded by this fragment. These sequences include heavy chain D segments, J segments and γ1 constant regions together with representatives of the 6 VH families (although this fragment is identified as the 670 kb fragment of HeLa cells from Berman et al. (1988), supra , we have discovered that this is an 830 kb fragment of human placenta and sperm DNA). Fractions containing this NotI fragment (see Figure 4) are ligated into the NotI cloning site of the pYACNN vector as described (McCormick, M. et al . (1990), Technique 2 , 65-71). Plasmid pYACNN is prepared by digestion of pYACneo (Clontech) with EcoRI and ligation in the presence of oligonucleotide 5'-AAT TGC GGC CGC-3 '

YAC clones containing the heavy chain NotI fragment are isolated as described in Traver et al . (1989), Proc. Natl Acad. Sci. USA , 86 , 5898-5902. The cloned NotI insert is isolated from high molecular weight yeast DNA by pulsed field gel electrophoresis as described in M. Finney, op. cit. The DNA is condensed by the addition of 1 mM spermine and microinjected directly into the nucleus of single cell embryos described previously. Alternatively, the DNA is isolated by pulsed field gel electrophoresis and introduced into ES cells by lipofection (Gnirke et al . (1991), EMBO J. , 10 , 1629-1634), or the YAC is introduced into cells ES by fusion of spheroplasts.

Example 20 Heavy Chain Transgen of Genomic Discontinuous Ig

An 85 kb SpeI fragment of human genomic DNA, containing VH 6, segments D, segments J, the constant region µ and part of the constant region γ (see Figure 4), has been isolated by cloning in YAC essentially as described in example 1. A YAC carrying a fragment of the germline variable region is isolated as described, such as a 570 kb NotI fragment upstream of the 670-830 kb NotI fragment described above containing multiple copies of V1 to V5. (Berman et al . (1988), supra . Detected two 570 kb Notl fragments, each containing multiple V segments.) The two fragments are injected together into the nucleus of a single mouse cell embryo as described in Example 1

Typically, the joint injection of two different DNA fragments results in the integration of both fragments at the same insertion site on the chromosome. For the Thus, approximately 50% of transgenic animals resulting that contain at least one copy of each of the two fragments will have the segment V segment fragment inserted above the fragment that contains the constant region. Of these animals, approximately 50% will perform the union of V to DJ by DNA inversion and approximately 50% per deletion, depending of the orientation of the 570 kb NotI fragment in relation to the 85 kb SpeI fragment position. DNA of animals is isolated resulting transgenic and, in animals in which it is discovered by Southern blot hybridization containing both transgenes (especially animals that contain both multiple human V segments as constant region genes human), the ability to express molecules of Human immunoglobulin according to conventional techniques.

Example 21 YAC Overlapping Fragment Union

Two YACs that carry an overlapping region are joined in a yeast by meiotic recombination as described in Silverman et al . (1990), Proc. Natl Acad. Sci. USA , 87 , 9913-9917, to obtain sequences that carry a single large YAC from the two smaller YACs. The two YACs align with respect to the arms, so that the attached YAC will contain an arm of the centromeric vector and a spleen of the non-centromeric vector. If necessary, the insert is cloned back into the vector using unique restriction sites at the ends of the insert. If the insert is not a single restriction fragment, unique sites are inserted into the arms of the vector by oligonucleotide transformation of yeasts, as described in Guthrie and Fink, op. cit. For YAC binding that carry non-contiguous sequences that do not overlap, an overlap is created as follows. The 3 'terminal region of the YAC 5' and the 5 'terminal region of the YAC 3' are subcloned, joined in vitro to create a binding fragment and reintroduced into one or both YACs by homologous recombination (Guthrie and Fink, op. cit.). Then, the two YACs are meiotically recombined as described in Silverman et al ., Op. cit. The bound YACs are introduced into mice, for example, as in example 1.

Example 22 Light Chain Transgen \ kappa of Genomic Human Ig

In Lorenz W. et al . (1987), Nucl. Acids Res ., 15 , 9667-9677 has described a map of the human κ light chain shown in Figure 11. A 450 kb XhoI fragment is isolated that includes the entire C κ, the enhancer 3 ', all J segments and at least five different V segments (a), or a 750 kb MluI to Notl fragment that includes all of the above plus at least 20 more V segments (b) and is introduced into zygotes or in ES cells as described in example 1.

Example 23 Light Chain Transgen \ kappa of Genomic Human Ig Formed by In Vivo Homologous Recombination

The 750 kb MluI to NotI fragment is digested with BssHII to produce a fragment of approximately 400 kb (c). The XhoI to NotI fragment of 450 kb (a) and the MluI fragment to BssHII of approximately 400 kb (c) have an overlapping sequence defined by the restriction sites BssHII and XhoI shown in the Figure 11. Homologous recombination of these two fragments after microinjection in a mouse zygote results in a transgene that contains at least an additional 15-20 V segments more than those found in the 450 kb XhoI / NotI fragment (example 22).

Example 24 Functionally variable region sequence identification reorganized into transgenic B cells

An antigen of interest is used to immunize (see Harlow and Lane, Antibodies: A Laboratory Manual , Cold Spring Harbor, New York (1988)) a mouse with the following genetic traits: homozygosis in the endogenous heavy chain locus for a deletion of J_ {examples 9 and 12); hemicigosis for a single transgen copy of reorganized human heavy chain minilocus (examples 5 and 14); and hemicigosis for a single copy of a reorganized human kappa light chain transgene (examples 7 and 16).

After the immunization program, the spleen and splenic cells are used to generate hybridomas. Be use cells from an individual hybridoma clone that secretes antibodies reactive with the antigen of interest to prepare DNA genomic A sample of genomic DNA is digested with several enzymes different constraints that recognize unique sequences of six base pairs and fractionated on an agarose gel. It's used Southern blot hybridization to identify two DNA fragments in the range of 2-10 kb, one of which contains the single copy of the string VDJ sequences heavy human reorganized and the other contains the single copy of the reorganized human light chain VJ sequence. These two fragments are fractionated by agarose gel size and cloned directly in pUC18. Then, the cloned inserts are then subclone respectively in heavy chain expression cassettes and lightweight that contain constant region sequences.

Plasmid clone p γ1 (example 14) is used as a heavy chain expression cassette and the rearranged VDJ sequences are cloned into the XhoI site. The plasmid clone pCK1 is used as a light chain expression cassette and the reorganized VJ sequences are cloned into the XhoI site. The resulting clones are used together to transfect SP 0 cells to produce antibodies that react with the antigen of interest (MS Co et al . (1991) Proc. Natl. Acad. Sci. USA 88 : 2869).

Alternatively, mRNA is isolated from the cloned hybridoma cells described above, and used to synthesize cDNA. Then, the expressed VDJ and VJ human heavy and light chain sequences are amplified by PCR and cloned (JW Larrich et al . (1989) Biol. Technology , 7 : 934-938). After determining the nucleotide sequence of these clones, oligonucleotides encoding the same polypeptides are synthesized and synthetic expression vectors are generated as described in C. Queen et al . (1989) Proc. Natl Acad. Sci. USA ., 84 : 5454-5458.

Therefore, it will be obvious that according to the invention a transgenic mouse is provided that has a genome that includes a transgene comprising DNA sequences encoding the variable (V), diversity (D), binding (J) regions and human constant of a human Ig protein, said sequences being operably linked to transcription regulatory sequences and capable of undergoing gene reorganization in vivo , when integrated into a non-human transgenic animal, to produce a reorganized gene encoding a polypeptide of human heavy chain, said construction also comprising a 5 'donor region of change µ from a constant region µ and a human acceptor region of change γ between the constant region µ and a constant region γ human, said said being change sequences operatively linked to effect a class change in vivo and for the production of heavy chain polypeptides ga Mma human.

Claims (1)

1. A transgenic mouse having a genome that includes a transgenic construct comprising DNA sequences encoding the human variable (V), diversity (D), binding (J) and constant regions of a human Ig protein, said said being sequences operably linked to transcriptional regulatory sequences and capable of undergoing gene reorganization in vivo , when integrated into a non-human transgenic animal, to produce a reorganized gene encoding a human heavy chain polypeptide, said construction also comprising a region 5 'change donor µ from a constant region µ and a human γ exchange acceptor region between the constant region µ and a human γ region, said change sequences being operatively linked to effect the change of in vivo class and production of human γ heavy chain polypeptides.